Advanced Therapy in Gastroenterology and Liver Disease - part 10 doc

One thousand two hundred and eleven patients from eight centers in Europe with pain and obstructive chronic pancreatitis underwent endoscopic therapy, including endoscopic pancreatic sph

782 / Advanced Therapy in Gastroenterology and Liver Disease

Since then there have been several studies that have

repro-duced these results, although a convincing benefit in

mor-tality has not been demonstrated (Imrie et al, 2002) Most

recently there have been reports that NG feeding is well

tol-erated in these patients Our practice is to begin enteral

feed-ing in patients by an endoscopically or radiologically placed

nasojejunal feeding tube after day 2 or 3 in patients with

severe pancreatitis Feeding is started at low rate of 20 cc/hr

Although this does not provide complete caloric

require-ments, small amounts of feeding are usually tolerated and

may preserve the intestinal barrier If nausea and vomiting

are present, a NG tube can be placed and kept to drainage

A small group of patients (between 10 to 20%) will not

tol-erate this method of feeding and require TPN Triglyceride

levels should be checked after the onset of feeding, especially

in patients with known hypertriglyceridemia

Pharmacotherapy with Cytokines, Enzyme Inhibitors,

and Anti-Inflammatory Agents

Multiple cytokines and anti-inflammatory mediators have

been implicated in the pathogenesis of acute pancreatitis

Thus, blockage by a single agent (eg, interleukin-10, lexipafant)

has not been effective in the treatment of pancreatitis

Surgical Therapy and Management of Other

Complications

Surgical therapy is used to treat infected necrosis or

com-plications The following chapter is on surgical

manage-ment of pancreatitis Surgical managemanage-ment includes

necrosectomy to remove necrotic tissue with intraoperative

and postoperative lavage of debris and pancreatic fluids

Percutaneous catheter-directed debridement of infected

necrosis has been described, but is best directed to patients

who are hemodynamically and clinical stable Further, a

number of these patients will subsequently require surgery

It is not our practice to recommend percutaneous drainage,

except in carefully selected patients with pancreatic

abscesses Endoscopic drainage with placement of

trans-gastric and tranduodenal catheters and nasopancreatic

tubes for irrigation has been described The experience

with this method is limited and it requires specific

endo-scopic experience, with experienced surgical backup Acute

fluid collections can be observed early in the course of acute

pancreatitis and will usually resolve spontaneously;

treat-ment is required only if they become symptomatic or

infected Pseudocysts are localized collections of necrotic

debris and fluid within a wall of granulation tissue that

per-sist for >4 weeks Regardless of its size, an asymptomatic

pseudocyst does not require any therapy and may be

observed indefinitely If the cyst is symptomatic (pain,

obstruction), it may be drained using a variety of methods

including endoscopic, percutaneous, and surgical Cysts

may be drained endoscopically either by cyst-gastrostomy

or cyst duodenostomy or by transampullary stent drainage.There is a separate chapter on endoscopic management

of pancreatitis (see Chapter 138, “Pancreatitis: EndoscopicTherapy”) Percutaneous drainage is attempted if the loca-tion precludes endoscopic cystgastrotomy or when mul-tiple noncommunicating cysts are observed If the

pseudocyst does not resolve with drainage, a proximal creatic duct stricture may be present and an ERCP should

pan-be performed ERCP prior to cyst drainage should only pan-be attempted if the pseudocyst can be immediately drained,

because if there is contrast entry into the pseudocyst (as incommunicating cysts) it increases the chance of infection

in the event that drainage is not performed Pseudocystsmay become infected, rupture, or hemorrhage

Post-ERCP Pancreatitis

Approximately 1 to 10% of ERCPs may be complicated

by acute pancreatitis This is usually mild, but severe creatitis and death may occur Table 136-5 summarizes riskfactors and preventive measures The judicious use of non-imaging procedures (magnetic resonance cholangiopan-creatography and EUS) to circumvent diagnostic ERCP isimportant because preventative measures have limited suc-cess Short duration (1 to 2 days) pancreatic duct stentingappears to be effective in high risk patients, including thoseundergoing pancreatic sphincterotomy, SO dysfunction,and pancreatic endotherapy

pan-Prevention of Recurrence

In patients with predicted biliary pancreatitis, if gallstonesare not observed on an initial ultrasound, the procedureshould be repeated after the pancreatitis has resolved.Patients with biliary pancreatitis have a significant risk ofrecurrence unless a cholecystectomy is performed Inpatients with mild to moderate pancreatitis, this can beaccomplished prior to discharge Patients who refusesurgery or are very high operative risks benefit from endo-scopic biliary sphincterotomy to prevent recurrence.Patients with alcohol use should be counseled about ces-

sation The treatment of patients with pancreas divisum is

controversial, but those with recurrent pancreatitis in whom

no other etiology is identified may benefit from

pancre-atic sphincterotomy of the minor ampulla Hyperlipidemic

patients should receive appropriate lipid lowering tions and patients with acute recurrent pancreatitis maybenefit from further endoscopic workup including ERCP,EUS, and SO manometry Ampullary or pancreatic cancermay present in pancreatitis Young patients or those with afamily history should be offered genetic testing in the appro-priate setting

OSCARJOEHINES, MD,AND HOWARDA REBER, MD

depression that often accompanies the chronic pain state.Nutrition may be impaired because oral intake is limited

by the pain that it produces The patient may be addicted

to narcotics

Type of Surgical Procedure

The type of operation depends on the anatomy of the creatic ductal system, and whether or not the pancreas isdiffusely involved with the disease or it involves one part

pan-of the gland more than the others Thus, patients with adilated pancreatic duct (>7 mm in the body of the gland)are candidates for a drainage operation that decompressesthe duct Those with a duct that is of normal caliber willprobably require resection of a part of the pancreas, usu-ally the head of the gland

Morbidity

Patients who undergo a pancreatic resection may develop

exocrine insufficiency and/or diabetes if enough pancreas is

removed Whereas this may be an acceptable price to payfor pain relief in some patients, others might be unable tomanage the dietary and insulin requirements that wouldensue (eg, patients who are addicted to narcotics and/oralcohol) So a resection operation might be contraindicated

in them However, even in patients with narcotic and/oralcohol addiction and a dilated duct, a duct decompressionoperation may be appropriate, because it almost never pro-duces exocrine or endocrine insufficiency Preoperativepsychiatric examination may help the surgeon to decideabout whether or not operation should be considered

Preoperative Imaging

All patients require preoperative imaging studies to helpwith decisions about the kind of operation that may be indi-cated The goals of imaging are (1) to assess the diameter ofthe duct, (2) to determine the presence of any associateddisease (eg, cysts, bile duct obstruction), and (3) to searchfor an unsuspected pancreatic malignancy All patientsshould undergo a high resolution computed tomography(CT) scan with fine cuts through the pancreas during thearterial phase of the study Endoscopic ultrasound (EUS)with fine needle aspiration (FNA) of any suspicious area

Chronic pancreatitis is characterized by chronic

inflam-mation with fibrosis and obliteration of both the endocrine

and exocrine components of the gland These changes are

irreversible, progressive, and may culminate in clinically

significant pancreatic insufficiency and diabetes Although

chronic alcoholism is the usual etiology, some patients

develop the disease because of chronic ductal obstruction,

some because of genetic predisposition, and a substantial

number for reasons as yet unknown Regardless of the

cause, the most common symptom of the disease is chronic

abdominal pain, and pain relief is the most frequent

rea-son for surgical intervention Other rearea-sons include

vari-ous intra-abdominal complications of pancreatitis (eg, bile

duct or duodenal obstruction, pseudocyst), and the

con-cern that pancreatic cancer may be present We will

describe the surgical considerations related to each

Chronic Abdominal Pain

The etiology of the pain is multifactorial and, in general,

is not well understood Factors include continued alcohol

consumption that results in local release of oxygen-derived

free radicals, diminished pancreatic blood flow and tissue

acidosis, perineural sheath destruction with exposure to

various nociceptive agents, and elevated pancreatic

duc-tal and parenchymal pressures The initial therapy for pain

in all of these patients should be nonoperative and includes

recommendations for the cessation of alcohol intake, and

the administration of oral analgesic agents

Referral for consideration of surgery requires (1) an

assessment of the significance of the pain for the individual

patient, which is highly subjective, (2) a determination of

the type of surgical procedure that might be appropriate,

and (3) an evaluation of the ability of the patient to deal

with any long term morbidity that an operation might

pro-duce

PAINASSESSMENT

In general, operation may be indicated in patients whose

pain interferes with the quality of their lives For

exam-ple, the attacks of pain may require frequent

hospitaliza-tions that interfere with school or employment The patient

may be unable to function productively because of the

Chronic Pancreatitis: Surgical Considerations / 785

may be indicated if the results of the CT scan raise a

ques-tion about malignancy There is a separate chapter on EUS

and FNA (see Chapter 5,“Endoscopic Ultrasonography and

Fine-Needle Aspiration”)

Operations to relieve pain in these patients either (1)

drain a dilated pancreatic ductal system or (2) resect

dis-eased pancreatic tissue if the duct is not enlarged The main

pancreatic duct normally measures 4 to 5 mm in the head,

3 to 4 mm in the body, and 2 to 3 mm in the tail of the

pan-creas The duct is considered dilated when it is at least

7 mm in diameter in the body of the gland

Ductal Drainage Operation

In patients with a dilated pancreatic duct, a ductal drainage

operation (longitudinal pancreaticojejunostomy, Puestow

procedure) is likely to be effective The operation involves

wide exposure of the anterior surface of the pancreas, and

incision of the anterior wall of the duct throughout the

length of the gland Pancreatic duct stones are removed and

a Roux-en-Y jejunal limb is constructed and sewn along

the length of the duct, so that pancreatic secretions empty

directly into the bowel

The operative results for longitudinal

pancreaticoje-junostomy are summarized in Table 137-1 The morbidity

and mortality (<2%) rates are minimal and there is almost

no risk of diabetes because little if any pancreatic tissue isresected Pain is relieved in 85% of patients for the first sev-eral years Most patients gain weight because they no longerexperience pain with eating, although the degree of mal-absorption does not change The major drawback of thisoperation is that within 5 years, pain recurs in as many as

40 to 50% of patients In a small number, this may be due

to stricturing of the anastomosis, but in most, it is bly associated with disease progression or the development

proba-of a complication Recurrence proba-of pain may also herald theappearance of pancreatic cancer

Pancreatic Resection

Patients with a normal diameter or narrowed duct may be

candidates for pancreatic resection This is especially true

when the pancreatic head is enlarged and contains multiplecysts and calcifications Pancreaticoduodenectomy(Whipple resection) or pylorus-preserving pancreatico-duodenectomy are performed most commonly, and we pre-fer the latter Pylorus preservation is felt by many to allowfor better postoperative nutrition and weight gain, but lit-tle objective data support this The operative mortality rate

is <3% and permanent pain relief is to be expected in 85

to 90% These operations are more likely to produceendocrine (22%) and exocrine (55%) insufficiency, which

is their major drawback Of course, some patients developthese problems anyhow as the disease progresses

In an effort to design an operation that would providepermanent pain relief and avoid the exocrine andendocrine insufficiency of a major resection, surgeons have

designed several new procedures that combine limited resection of the head of the pancreas with a pancreaticoje- junostomy The so-called Beger or Frey operations remove

most of the head of the pancreas except for a shell of creatic tissue posteriorly The cavity thus created is drainedinto a Roux-en-Y limb of jejunum; gastroduodenal con-tinuity is not disturbed This operation can be performedwhether or not the pancreatic duct is dilated If it is, the

pan-TABLE 137-1 Results of Longitudinal

Pancreaticojejunostomy for Chronic Pancreatitis

TABLE 137-2 Results of Resection for Chronic Pancreatitis

pan-pancreaticojejunostomy is extended over the body of the

pancreas to incorporate the dilated duct in that area Early

results suggest that pain relief is excellent in 85 to 90% of

patients, that the relief persists beyond several years, and

that exocrine or endocrine insufficiency are not

precipi-tated by the surgery In those patients whose bile duct has

also been obstructed by the fibrotic pancreas, this “coring”

of pancreatic tissue from the head of the gland may

decom-press that duct as well This operation is contraindicated if

there is a concern about the presence of a malignant

neo-plasm in the head of the pancreas; a

pancreaticoduo-denectomy should be performed in these cases

Uncommonly, chronic pancreatitis is localized

pre-dominantly in the body or tail of the pancreas In these

cases, a distal pancreatectomy (with or without

splenec-tomy) may be effective The surgeon should investigate the

possibility that an occult malignancy may have produced

a more proximal duct obstruction, and that a neoplastic

duct stricture is the reason for such localized pancreatitis

Otherwise in patients with predominantly distal disease,

distal pancreatectomy is safe and pain relief can be expected

in as many as 90% of patients after 4 years For the usual

patient who has diffuse disease involving the entire

pan-creas, distal pancreatectomy is ineffective, however Because

it results in a brittle diabetes which is often difficult to

con-trol, and because lesser procedures are likely to be effective,

total pancreatectomy for chronic pancreatitis is almost

never done today

Complications of Chronic Pancreatitis

Pseudocyst

A pancreatic pseudocyst is a collection of fluid usually in

the vicinity of the pancreas, which develops in association

with a leak of pancreatic juice from the inflamed

parenchyma or from a disrupted duct The wall of the

pseudocyst is comprised of fibrous nonepithelialized

tis-sue Occasionally a pseudocyst may present at great

dis-tance from the pancreas (eg, thorax, groin), when the fluid

dissects through tissue planes The majority of acute

pseudocysts that appear during an episode of acute

pan-creatic inflammation resolve without intervention

However, most pseudocysts that develop on a background

of chronic pancreatitis are unlikely to resolve

sponta-neously, and they may need treatment Asymptomatic

pseudocysts up to 5 to 6 cm in diameter may be safely

observed, and are usually followed with either serial

ultra-sound or CT examinations Larger cysts or pseudocysts of

any size that are symptomatic require intervention

Symptoms are most often from gastrointestinal (GI)

obstruction when the cyst distorts or compresses the

stom-ach, duodenum, or bile duct, or produces abdominal pain

Serious complications also can occur, although they are

uncommon (<5% of cases) These include hemorrhage

into the cyst, perforation of the cyst, or cyst infection.Hemorrhage is usually caused by erosion of the splenic orgastroduodenal artery or other major vessel within the wall

of the cyst, and the bleeding is often confined to the cystlumen The diagnosis should be suspected if there are clin-ical signs of hypovolemia and a falling hematocrit Theremay be abdominal pain, and a mass may be palpable Anabdominal CT scan shows the cyst with the containedblood clot Angiography confirms the diagnosis, and theradiologist should attempt to embolize the bleeding ves-sel If not, emergency surgery with ligation of the vessel

or excision of the cyst is required Perforation of a cyst is a surgical emergency that is characterized by the sud-den onset of intense abdominal pain with peritonitis.Patients require urgent surgery with irrigation of the peri-toneal cavity and usually external cyst drainage Infection

pseudo-of a pseudocyst should be suspected if signs pseudo-of sepsisdevelop Diagnosis by CT scan and treatment by percuta-neous cyst aspiration and drainage are usually effective

In the absence of a life threatening complication, tive surgery of pseudocysts is usually delayed until the cysthas developed a mature wall that will hold sutures at thetime of repair For those cysts that develop following anepisode of acute pancreatitis, this requires 4 to 6 weeks Inmost cases the patient can eat and be discharged from thehospital during the interval Pseudocysts that resolve spon-taneously usually will do so during this time If no episode

elec-of clinically significant acute pancreatitis preceded thedevelopment of the cyst, as is often the case in patientswith chronic pancreatitis, usually no waiting period is nec-essary

Pseudocysts may be treated surgically, or by endoscopic

or radiologic drainage Endoscopic methods require the

placement of a plastic stent through the stomach or denal wall into the adjacent cyst The stent is eventuallyremoved, and in about 80% of cases, the cyst is perma-nently eradicated These endoscopic techniques requireexpertise, which still is not widely available They are dis-cussed in the next chapter (see Chapter 138, “Pancreatitis:Endoscopic Therapy”) Radiologic approaches usually con-sist of percutaneous external drainage of the cyst with even-tual removal of the drainage catheter many weeks later.Many of these pseudocysts recur Surgical treatment usu-ally consists of drainage of the cyst internally to either thestomach (cystgastrostomy) or to a Roux-en-Y limb ofjejunum (cystjejunostomy) Both are safe and effective,with recurrence rates <10% If the pseudocyst is in the tail

duo-of the pancreas, a distal pancreatectomy with excision duo-ofthe cyst may be best

Pancreatic Fistula

In the setting of chronic pancreatitis, a pancreatic fistula isusually the result of a ductal disruption from an episode ofacute pancreatitis The diagnosis is made by finding a high

amylase level (usually many thousands of U/L) in the fistula

effluent Some fistulas will close spontaneously, provided

that ductal continuity can be re-established as healing

occurs, infection is eradicated, and nutrition is adequate

However, the frequent presence of duct obstruction in

chronic pancreatitis may make it less likely that the

fis-tula will close Parenteral nutrition is usually not required

and most patients are able to eat a regular diet There is

no evidence that oral intake delays resolution of fistula

The use of somatostatin does not appear to hasten fistula

closure, although if it is a high output fistula (ie,>

200 mL/d), the secretory inhibitor may simplify

man-agement of the patient Fistulas that persist for as long as

1 year, or those whose anatomic characteristics preclude

spontaneous closure (eg, duct obstruction between

fis-tula and duodenal lumen, duct discontinuity), will require

operative repair This is best done by creating an

anasta-mosis between the pancreatic duct at the point of the leak

and a Roux-en-Y limb of jejunum The success rate of

operative repair is >90%

Biliary Stricture or Obstruction

Jaundice may occur in up to one-third of patients with

chronic calcific pancreatitis at some point during the

dis-ease, usually when there is pancreatic swelling at the time of

an episode of acute pancreatitis This often resolves as the

acute inflammation subsides, but as many as 10% of patients

are left with obstruction of the common duct This is due to

fibrosis of the head of the pancreas resulting in constriction

of the duct as it passes through this portion of the gland The

stricture usually appears as a long, symmetrical narrowing

when it is visualized by magnetic resonance

cholangiopan-creatography or endoscopic retrograde cholangiography

The proximal duct and gallbladder may be distended, but

obstruction of the duct is almost never complete, which

dif-ferentiates it from a malignant obstruction A simple biliary

bypass using a Roux-en-Y choledochojejunostomy

effec-tively treats such a biliary stricture Endoscopic procedures

are discussed in the next chapter

Intestinal Compression or Obstruction

A minority of patients will present with obstruction of the

second or third portion of the duodenum Upper endoscopy

and CT scan should be performed to rule out the presence

of a neoplastic process Then a loop gastrojejunostomy can

be done to bypass the obstruction

Obstruction of the colon (usually the transverse or

splenic flexure) can also occur from chronic pancreatitis If

this is due to an episode of acute inflammation, the

obstruc-tion will likely resolve If it persists, then a colonoscopy

should be performed to rule out malignancy Persistence of

the obstruction requires a resection of the involved segment

of colon and an end-to-end anastomosis

Pancreatic MalignancyPatients with long standing chronic pancreatitis are at a10% lifetime risk for the development of pancreatic ade-nocarcinoma During examination for surgery in apatient with chronic pancreatitis, imaging studies mayshow focal changes in the pancreas that suggest malig-nancy, or other aspects of the clinical presentation (eg,rising or markedly elevated CA19-9, change in character

of pain, accelerated weight loss) may have raised the ician’s index of suspicion about the possibility that can-cer is present

clin-If there is concern about malignancy, we recommendEUS examination, which is currently the most sensitive diag-nostic study to identify small cancers, and also can be used

to obtain tissue from the lesion that could confirm the nosis EUS and FNA are discussed in Chapter 5 (“EndoscopicUltrasonography and Fine-Needle Aspiration”)

diag-However, whether or not the diagnosis is confirmed operatively, patients in whom the surgeon suspects thecoexistence of pancreatic cancer with underlying chronicpancreatitis require pancreatic resection This means a pan-creaticoduodenectomy for head and uncinate lesions and

pre-a distpre-al ppre-ancrepre-atectomy for body pre-and tpre-ail lesions Even ifcancer is not found when the resected specimen is exam-ined by the pathologist, this approach represents the cur-rent standard of care in such circumstances This is becauseresection operations are safe, they are one of the standardoperations normally done for chronic pancreatitis withoutcoexisting cancer, and pancreatic cancer is uniformly fatalwhen it is not surgically resected

ConclusionThe surgical considerations for patients with chronic pan-creatitis include procedures to address chronic pain, vari-ous complications of the disease, and pancreatic cancer.The decision to operate in any single patient with painfrom the disease is complex, and should be based on a vari-ety of factors that include the psychosocial makeup of thepatient as well as pancreatic anatomy If surgery is indi-cated, the type of operation hinges on the appearance ofthe pancreatic ducts (Figure 137-1) In patients with adilated duct, a drainage procedure is often the best optionbecause this offers good pain relief and the least long termmorbidity If the ducts are not enlarged, failed prior duc-tal drainage, or there is concern about the presence of can-cer, then resection should be performed Newer operationsthat resect most of the head of the pancreas but still pre-serve GI continuity (Beger et al, 1999; Frey and Amikura,1994) may provide the best long term pain relief with theleast long term morbidity Patients with chronic pancre-atitis can develop pseudocysts, pancreatic fistulas, and bil-iary or intestinal obstruction The pancreatic surgeon must

be prepared to deal with all of these issues

Chronic Pancreatitis: Surgical Considerations / 787

FIGURE 138-1 (A) Technique of pancreatic sphincterotomy using a pull type sphincterotome (Left top) Biliary sphincterotomy is performed using

a standard pull type sphincterotome (Right top) Pancreatic spincterotomy performed with a pull type sphincterotome cutting in the 1 o’clock tion (Left bottom) Completed biliary and pancreatic sphincterotomy A guidewire is in the pancreatic duct (Right bottom) A 6 F pancreatic stent is

direc-placed following performance of the pancreatic sphincterotomy (B) Traction sphincterotome positioned in minor papilla Note the extent of the minor

papilla mound (arrows) Duodenal juice at the minor papilla orifice is aspirated away before cutting to prevent heat dissipation to juice and boiling

the adjacent tissues during the sphincterotomy (C) Wire is bowed taught and cut is performed rapidly with minimal coagulation utilizing the ERBE generator The optimal cut length in this setting is unknown The 5 mm length minor papilla sphincterotomy is complete without white tissue coag- ulum (D) White pancreatic stone removed through patent sphincterotomy orifice with balloon catheter

B A

Pancreatitis: Endoscopic Therapy / 791

ventral pancreatogram, a major papilla pancreatic

sphinc-terotomy is usually performed to facilitate access to the

duct prior to attempts at stricture dilation and stent

place-ment There are two methods to cut the major pancreatic

sphincter A standard pull-type sphincterotome (with or

without a guide wire) is inserted into the pancreatic duct

and oriented along the axis of the pancreatic duct (usually

in the 12 to 1 o’clock position) Although the landmarks

to determine the length of incision are imprecise,

author-ities recommend a cutting length of 5 to 10 mm The

cut-ting wire of the sphincterotome should not extend more

than 6 to 7 mm up the duct when applying electrocautery,

so as to prevent deep ductal injury Alternatively, a

needle-knife can be used to perform the sphincterotomy over a

previously placed pancreatic stent Performing a biliary

sphincterotomy first can expose the pancreaticobiliary

sep-tum and allow the length of the cut to be gauged more

accurately

Cannulation of the ventral pancreatic duct may be

unsuccessful in patients with recurrent, idiopathic

pan-creatitis and should heighten the suspicion that pancreas

divisum may be present If the pancreatogram from the

major papilla is successful, a fine branching of the ventral

duct typical of pancreas divisum must be ascertained In

patients over the age of 40 years with pancreatic duct

cut-off in the head of the pancreas, the endoscopist must be

wary of underlying pancreatic malignancy causing an

abrupt cut-off of the ventral pancreatic duct

masquerad-ing as pancreas divisum If the ventral pancreatogram is

typical of pancreas divisum, the minor papilla is sought

Approaches to the minor papilla for cannulation begin

with an endoscopic still photo of the minor papilla

ori-fice to refer to during attempted minor papilla cannulation

(Lehman and Sherman, 1995) The use of intravenous

secretin (Secreflo, Repligen Inc.) 16 µg to induce

pancre-atic ductal secretion may aid in identifying the minor

papilla orifice and successful cannulation We have reported

our experience with spraying the minor papilla with

meth-ylene blue prior to secretin administration to more

accu-rately locate the minor papilla and improve success of

minor papilla cannulation (Park et al, 2003) For

cannula-tion, we commonly use the highly tapered 3-4-5 F catheter

loaded with the 0.018 inch roadrunner guide wire

(Wilson-Cook, Winston-Salem, NC) Once the dorsal duct is

accessed and the anatomy is defined, minor papilla

ther-apy is considered We reserve minor papilla therther-apy for

patients who have experienced at least two bouts of

doc-umented pancreatitis or for patients with chronic

pancre-atitis with strictures, duct disruptions or stones in the

dorsal duct In patients with pancreas divisum, a minor

papilla sphincterotomy is usually necessary The minor

papilla sphincterotomy may be performed with a

needle-knife sphincterotome over a 3 or 4 F pancreatic stent or

with a standard pull-type sphincterotome The technique

is similar to that of major papilla pancreatic tomy, except that the direction of the incision is usually

sphinctero-in the 10 to 12 o’clock position and the length of thesphincterotomy is limited to 4 to 8 mm

Chronic Pancreatitis

Endoscopic therapy is now being applied in the setting ofchronic pancreatitis for patients presenting with painand/or clinical episodes of acute pancreatitis One of theaims of endoscopic therapy is to alleviate the obstruction

to exocrine juice flow Outflow obstruction may be caused

by ductal strictures (biliary or pancreatic), pancreaticstones, pseudocysts, and minor or major papilla stenosis.Although the endoscopic approach has never been directlycompared with surgery, endoscopic drainage is appealing

in that it may offer an alternative to surgical drainage cedures, with generally less morbidity and mortality.Furthermore, endoscopic procedures do not preclude sub-sequent surgery, should it be necessary Moreover, the out-come from reducing the intraductal pressure by endoscopicmethods may be a predictor for the success of surgicaldrainage

pro-Benign strictures of the main pancreatic duct may be aconsequence of generalized or focal inflammation ornecrosis around the main pancreatic duct Given the puta-tive role of ductal hypertension in the genesis of symptoms(at least in a subpopulation of patients), the utility of pan-creatic duct stents for treatment of dominant pancreaticduct strictures is being evaluated Underlying malignancymust be considered and excluded by tissue sampling atERCP or by endoscopic ultrasonography with fine-needleaspiration of the strictured segment Most pancreatic stentsare simply standard polyethylene biliary stents with extraside holes at approximately 1 cm intervals to permit betterside branch juice flow Stents made of other materials havereceived limited evaluation The technique for placing astent in the pancreatic duct is similar to that used for insert-ing a biliary stent In most patients, a pancreatic sphinc-terotomy (with or without a biliary sphincterotomy) viathe major or minor papilla is performed to facilitate place-ment of accessories and stents A guide wire must bemaneuvered upstream to the narrowing Hydrophilic flex-ible tip wires are especially helpful for bypassing strictures.Torqueable wires are occasionally necessary to achieve thisgoal High-grade strictures require dilatation prior to inser-tion of the endoprosthesis This may be performed withhydrostatic balloon dilating catheters or graduated dilat-ing catheters Extremely tight strictures may permit pas-sage of only a small caliber guide wire Such wires may beleft in situ overnight and usually permit dilator passage thenext day Alternatively, 3 F angioplasty balloons or theSoehendra stent retriever may be helpful to effectively dilatethe stricture Although one preliminary report suggestedthat luminal patency of the duct persisted at a mean time

of 5 months following balloon dilation alone, most

author-ities have observed recurrence of strictures after onetime

dilation and therefore advocate stenting As a rule, the

diameter of the stent should not exceed the size of the

downstream duct Therefore, 5, 7, or 8.5 F stents are

com-monly used in smaller ducts, whereas 10 to 11.5 F stents or

dual side-by-side 5 to 7 F stents may be inserted in patients

with severe chronic pancreatitis and a dilatated main

pan-creatic duct The tip of the stent in the pancreas must

extend upstream to the narrowed segment and into a

straight portion of the pancreatic duct to avoid stent tip

erosion through the duct wall For diagnostic trials of

pan-creatic stenting in patients with nearly daily pain, most

stents are left in place for 3 to 4 weeks

The appropriate duration of pancreatic stent placementand the interval from the placement to change of the pan-creatic stent is not known The following two options areavailable: (1) the stent can be left in place until symptoms

or complications occur and (2) the stent can be left in placefor a predetermined interval (eg, 3 months) If the patientfails to improve, the stent should be removed, because duc-tal hypertension is unlikely to be the cause of pain If thepatient has benefited from stenting, one can remove thestent and observe the patient clinically, continue stentingfor a more prolonged period, or perform a surgicaldrainage procedure

The majority of patients with a stricture have associatedcalcified pancreatic duct stones For optimal results, the

FIGURE 138-2 A 57-year-old male with history of alcohol-induced chronic pancreatitis with recurrent monthly attacks of pain requiring

multi-ple hospitalizations Pancreatogram via major papilla reveals tight stricture of downstream pancreatic duct with multimulti-ple filling defects in the

dilat-ed upstream pancreatic duct (A) A hydrophilic 0.025 inch guide wire is usdilat-ed for engagement of the stricture and access to the dilatdilat-ed duct (B) A

4 mm diameter hydrostatic balloon catheter is used for stricture dilatation after a 5-7-10 F dilation catheter was not successfully passed through the stricture (C) One 7 F by 8 cm long with a three-quarter external pigtail and single internal flap was placed into the pancreatic duct This stent was removed 1 month later and the patient has been without pain 1 year later (D).

A

C

B

D

Pancreatitis: Endoscopic Therapy / 793

therapy must address both the stones and stricture In the

largest multicenter trial, Rosch and colleagues (2002)

reported on the long-term follow up of over 1000 patients

with chronic pancreatitis undergoing initial endoscopic

therapy during the period 1989 to 1995 One thousand two

hundred and eleven patients from eight centers in Europe

with pain and obstructive chronic pancreatitis underwent

endoscopic therapy, including endoscopic pancreatic

sphincterotomy, pancreatic stricture dilation, pancreatic

stone fragmentation by extracorporeal shock wave

lithotripsy and pancreatic stone removal, pancreatic stent

placement, or a combination of these methods One

thou-sand eighteen patients (84%) were observed for

sympto-matic improvement and need for pancreatic surgery over

a mean of 4.9 years (range 2 to 12 years) Success of

endo-scopic therapy was defined as a significant reduction or

elimination of pain and reduction in pain medication

Partial success was defined as reduction in pain, though

further interventions were necessary for pain relief Failure

of endoscopic therapy was defined as the need for

pan-creatic decompressive surgery or patients that were lost

to follow-up Over long-term follow-up, 69% of patients

were successfully treated with endoscopic therapy and 15%experienced a partial success Twenty percent of patientsrequired surgery, with a 55% significant reduction in pain.Five percent of patients were lost to follow-up The group

of patients that had the highest frequency of completedtreatment were patients with stones alone (76%) as com-pared to patients with strictures alone (57%) and patients

with strictures and stones (57%) (p<.001) Interestingly,the percentage of patients with no or minimal residual pain

at follow-up was similar in all groups (strictures alone 84%,

stones alone 84%, and strictures plus stones 87%) (p =

.677) The authors of this report concluded that endoscopictherapy for chronic pancreatitis in experienced centers iseffective in the majority of patients and the beneficialresponse to successful endoscopic therapy in chronic pan-creatitis is durable and long term

Endoscopic methods alone will likely fail in the ence of large or impacted stones and stones proximal to astricture Extracorporeal shock wave lithotripsy can be used

pres-to fragment spres-tones and facilitate their removal Thus, thisprocedure is complementary to endoscopic techniques andimproves the success of nonsurgical ductal decompression

FIGURE 138-3 A 40-year-old female with alcohol-induced chronic pancreatitis complicated by pancreatic main duct stones Pancreatogram

revealing dilatated pancreatic duct with 5 mm diameter filling defect consistent with a pancreatic stone (A) After pancreatic sphincterotomy, a wire guided stone extraction basket was used The basket is opened fully in the dilatated pancreatic duct and the stone is engaged (B) Basket is slowly closed on the stone (C) Stone is extracted and follow-up pancreatogram with a balloon catheter reveals no residual filling defects No fur- ther stenting was performed (D).

non-D C

B A

In a recent retrospective review of the efficacy of ESWL as

an adjunct to endoscopic therapy, Kozarek and colleagues

(2002) examined 40 patients who underwent a total of 46

ESWL sessions (average 1.15 sessions per patient) Eighty

percent (80%) of patients did not require surgery and had

significant pain relief, reduced number of hospitalizations,

and reduced narcotic use as compared to the pre-ESWL

period over a mean 2.4 years follow-up

Only randomized controlled studies comparing gical, medical, and endoscopic techniques will allow us

sur-to determine the true long-term efficacy of pancreaticduct stenting for stricture therapy There remain manyunanswered questions: Which patients are the best can-didates? Is proximal pancreatic ductal dilatation a pre-requisite? Does the response to stenting depend on theetiology of the chronic pancreatitis? Finally, as noted, how

FIGURE 138-4 A 41-year-old woman with a history of abdominal pain, pancreatitis and pancreatic calcification on computed tomography scan.

Abdominal radiograph reveals solitary radio-opaque stone in head/body region (A) Pancreatogram reveals an 8 mm obstructing stone in body of creas pancreatic duct (B) A 0.018 inch diameter guide wire was advanced beyond the stone Further contrast filling of duct demonstrating upstream dilatation Stone extraction with basket was unsuccessful (C) Extracorporeal shock wave lithotripsy performed with excellent fragmentation of stone Pancreatogram 1 week post-ESWL Mild duct irregularity in body of pancreas duct with minimal upstream dilatation All stone fragments were removed and no pancreatic stent was placed (Patient reported a history of abdominal injury in an auto accident over 10 years before the onset of symptoms) (D).

pan-A

D C

B

Pancreatitis: Endoscopic Therapy / 795

does endoscopic therapy compare with medical and

sur-gical management?

Pancreatic Pseudocysts

Pancreatic pseudocysts as complications of acute and

chronic pancreatitis are being tackled more frequently by

experienced endoscopists (Lehman, 1999) A word of

cau-tion if you as the endoscopist consider wading into

pseudo-cyst territory: Obtain and review a pre-ERCP high quality

pancreas CT scan, become very familiar with transpapillary

and transluminal drainage techniques, have experienced

interventional radiology and surgical back-up, and have an

understanding of potential complications A high quality CT

scan of the pancreas and surrounding structures will define

the proximity of the pseudocyst to the luminal wall (<10

mm thickness between lumen and pseudocyst is optimal),

contents of pseudocyst, the presence of splenic artery

aneurysm, splenic thrombosis and/or gastric varices, and the

extent of pancreatic necrosis At the initial ERCP, a

pancre-atogram is attempted to evaluate whether duct

communi-cation is present Transpapillary drainage of pseudocysts may

be performed by placing the stent either into the duct

upstream from the communication (“bridging the

disrup-tion”) or placing the stent into the pseudocyst cavity itself

Pancreatic sphincterotomy is commonly performed to ablate

the sphincter pressure gradient, and downstream pancreatic

strictures, if present, are dilated with balloon or passage

catheter dilators In a large series of endoscopic therapy for

acute and chronic pancreatic pseudocysts by Baron and

col-leagues (2002), 63% of 136 patients had communication

of the collection, with the main pancreatic duct affording

the potential for transpapillary therapy

Transluminal drainage via transduodenal or

transgas-tric approach may be the preferred technique in patients

with pancreatic duct cut-off, noncommunicating

pseudo-cysts, large tail of pancreas pseudocysts Endoscopy with

the therapeutic duodenoscope is performed to assess for

a bulge on the gastric wall or duodenal wall Once

local-ized, sampling or injection of the pseudocyst with the

Howell aspiration needle (Wilson-Cook Inc,

Winston-Salem, NC) may be performed prior to needle-knife

punc-ture into the pseudocyst The needle-knife puncpunc-ture is

performed perpendicular to the lumen with short bursts

of cautery and a forceful entry into the pseudocyst cavity

This perpendicular approach will minimize tracking into

the gut wall and reduce bleeding A guide wire is placed

into the pseudocyst and coiled in the cavity to maintain

access Balloon catheter dilation of the fistulous tract is

per-formed and single or multiple double-pigtail stents are

placed Technical success with proper patient selection

approaches 90% in most series In Baron’s series, complete

endoscopic resolution was achieved in 113 of 138 patients

Patients with chronic pseudocysts obtained the best result

(59 of 64 resolved, 92%) as compared to patients with acute

pseudocysts (23 of 31, 74%) or necrosis (31 of 43, 72%).Complications of transluminal endoscopic drainage ofpseudocysts, including bleeding, infection, and perfora-tion, total 20% in combined series Death has beenreported in 1% of over 500 patients undergoing endoscopicdrainage of pseudocysts, similar to the reported mortality

of surgically treated patients

ConclusionEndoscopic therapy for acute pancreatitis, chronic pan-creatitis and pancreatic pseudocysts has continued toevolve with improved innovative techniques and betterpatient selection Results of endoscopic therapy for pan-creatic diseases, in experienced hands, rivals the resultsobtained with surgery Early ERCP in the setting of severeacute gallstone pancreatitis is routinely performed in aca-demic and community practices with success Endoscopictherapy for chronic pancreatitis with stricture dilatation,pancreatic stone extraction, and pancreatic sphincterotomyafford short- to medium-term relief in the majority ofpatients Endoscopic pseudocyst drainage is the most tech-nically demanding procedure in managing patients withpancreatic disease, but with proper patient selection, goodresults may be achieved by experienced endoscopists in themajority of patients

Supplemental Reading

Baron TH, Harewood GC, Morgan DE, et al Outcome differences after endoscopic drainage of pancreatic necrosis, acute pan- creatic pseudocysts, and chronic pancreatic pseudocysts Gastrointest Endosc 2002;56:7–17.

Fogel EL, Sherman S Acute biliary pancreatitis: When should the endoscopist intervene? Gastroenterology 2003;125:229–35 Kozarek RA, Brandabur JJ, Ball TJ, et al Clinical outcomes in patients who undergo extracorporeal shock wave lithotripsy for chron-

ic calcific pancreatitis Gastrointest Endosc 2002;56:496–500 Lehman GA Pseudocysts Gastrointest Endosc 1999;49(Part 2):S81–4.

Lehman GA, Sherman S Pancreas divisum: diagnosis, clinical nificance, and management alternatives Gastrointest Endosc Clinics N America 1995;5:145–70.

sig-Neoptolemos JP, Carr-Locke DL, London NJ, et al Controlled trial

of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones Lancet 1988;2:979–83 Park S-H, de Bellis M, McHenry L, et al Use of methylene blue

to identify the minor papilla or its orifice in patients with creas divisum Gastrointest Endosc 2003;57:358–63 Rosch T, Daniel S, Scholz M, et al Endoscopic treatment of chron-

pan-ic pancreatitis: a multpan-icenter study of 1000 patients with term follow-up Endoscopy 2002;34:765–71.

long-Tarnasky PR, Hawes RH Endoscopic diagnosis and therapy of unexplained (idiopathic) acute pancreatitis Gastrointest Endosc Clinics N America 1998;8:13–34.

PETERDRAGANOV, MD,AND PHILLIPP TOSKES, MD

exocrine function may deteriorate even if the patient stopsdrinking That progression of exocrine dysfunction doesnot occur in patients with small duct disease and theirproblem continues to be pain and not steatorrhea.Approximately 20% of patients with chronic pancreatitismay present with endocrine or exocrine dysfunction in theabsence of abdominal pain (Layer et al, 1995) Progression

to end-stage pancreatitis is associated with significant bidity, mortality, and utilization of societal resources

mor-Pathogenesis

Alcohol abuse continues to be the most common cause of

chronic pancreatitis worldwide, accounting for 60 to 70%

of all patients with chronic pancreatitis In the 30 to 40%

of subjects with chronic pancreatitis where the etiology is

not due to alcohol, the most common setting is that of pathic pancreatitis, which may account for 20% of the cases.

idio-Table 139-1 lists the causes of chronic pancreatitis, and it is

important to point out that cystic fibrosis (CF) is the most

common cause in children There is a separate chapter oncystic fibrosis (see Chapter 140, “Cystic Fibrosis and OtherHereditary Diseases of the Pancreas”) As more knowledgeabout the pathophysiology of chronic pancreatitis is gained,

it is becoming appreciated that there are several subsets ofpatients with chronic pancreatitis It is likely that with the

recognition that a cholecystokinin (CCK)-releasing factor is

important in the feedback control of pancreatic exocrinesecretion, an appreciation is emerging that there may be

OverviewPatients with chronic pancreatitis come to medical atten-

tion because of abdominal pain and/or maldigestion The

management of patients with maldigestion (steatorrhea)

is relatively easy with the use of potent pancreatic enzymes

The management of abdominal pain continues to be a

frus-tration for both the patient and the physician managing

that patient The pathogenesis of the pain associated with

chronic pancreatitis is ill understood Clinicians must

appreciate that chronic pancreatitis is not one disease, but

that there are subsets of patients labeled with chronic

pan-creatitis who must be approached based on the

patho-physiology present It is helpful to divide patients with

chronic pancreatitis into those with big duct disease and

those with small duct disease The diagnostic and

thera-peutic approaches to patients with chronic pancreatitis are

greatly influenced by the degree of damage to the exocrine

pancreas The cornerstone of medical management for

either abdominal pain or maldigestion is the employment

of pancreatic enzyme formulations If severe damage to the

pancreas is present, endoscopic or surgical decompression

of the main pancreatic duct or suppression of pancreatic

secre-tion with agents such as octreotide may afford pain relief.

Presentation

Patients with chronic pancreatitis seek medical attention

because of abdominal pain or symptoms of maldigestion

(chronic diarrhea, steatorrhea, weight loss, and fatigue)

The abdominal pain is quite variable in location, severity,

and frequency The pain of chronic pancreatitis often does

not follow a classical pattern, such as that seen in acute

pan-creatitis where the pain is mid-epigastric and radiates to

the back Indeed, chronic pancreatitis pain may mimic

other causes of abdominal pain and may be quite

nonspe-cific in its clinical presentation The pain can be constant

or intermittent and there may be frequent pain-free

inter-vals Eating may exacerbate the pain, leading to fear of

eat-ing and weight loss The spectrum of abdominal pain

ranges from mild to severe and narcotic addiction is often

a frequent complication of chronic pancreatitis pain

In those patients with abdominal pain secondary to big

duct disease, often as a result of alcohol abuse, the patient’s

TABLE 139-1 Causes of Chronic Pancreatitis

Alcohol-induced (60%) Idiopathic (20%) Other (20%)

Chronic Pancreatitis / 797

forms of pancreatitis that are related to abnormalities in

CCK homeostasis (Chey, 1997) This CCK-releasing factor

has been isolated and sequenced, and it may be available

as a diagnostic test in the very near future One issue that

must be faced is that clinicians have attempted to treat all

patients with chronic pancreatitis in a similar fashion, which

has generally not been successful It is helpful to begin to

look at chronic pancreatitis subsets by dividing the patients

into those with big duct disease and those with small duct

disease The diagnostic approach, clinical course, and

ther-apeutic management are quite different in patients with

these two kinds of chronic pancreatitis These differences

will be stressed throughout the discussion

A great deal of attention has been paid recently to the

finding of whether or not genetic defects can account for

the majority of patients with idiopathic chronic

pancre-atitis Despite enthusiastic expectations, it is becoming

quite clear that most cases of idiopathic chronic

pancre-atitis represent a complex interaction of genetic and

envi-ronmental factors rather than resulting from a single gene

mutation Indeed at the present time, it appears that most

experts would agree that genetic abnormalities may account

for approximately 15% of those patients in the idiopathic

category Optimistic clinicians among us had thought that

this percentage would be much CF fibrosis conductance

regulator (CFTR) gene account for the chronic

pancreati-tis noted in most patients with idiopathic chronic

pancre-atitis Other genetic mutations that may be important in

chronic pancreatitis include the secretory trypsin inhibitor

gene (SPINK1) and polymorphisms in genes modulating

inflammation such as interleukin (IL)-1, IL-10, and tumor

necrosis factor (TNF) CFTR mutations are common in the

general population It is believed that the pancreatitis risk

is increased approximately 40-fold by having two CFTR

mutations, 20-fold by having a SPINK1 mutation, and

900-fold by having both two CFTR mutations and a SPINK1

mutation (Grendell, 2003).*

DiagnosisThe diagnosis of chronic pancreatitis is rather easy in

patients with big duct disease and quite challenging in

patients with small duct disease Table 139-2 lists diagnostic

tests that are carried out worldwide in the examination of

patients with chronic pancreatitis Tests of function usually

precede tests of structure in regard to sensitivity The most

sensitive and most specific tests are at the top of each

col-umn with the least sensitive and specific at the bottom In

2004, the most appropriate way of detecting whether or not

chronic pancreatitis is present is a combination of a secretin

hormone stimulation test and endoscopic ultrasonography

(EUS) In big duct disease, just about any test chosen will be abnormal In our experience, we have used a serum trypsino- gen as a first line test in patients who appear to have severe

pancreatic insufficiency (PI) and present with tion/maldigestion That test is now commercially availablefrom general diagnostic laboratories In our laboratory, thenormal values are 29 to 58 ng/mL Patients with chronic pan-creatitis of a mild to moderate degree have values of 20 to

malabsorp-28 ng/mL; patients with values <20 ng/mL have severe PIand this value usually correlates well with the presence ofsteatorrhea A new test that has been introduced into theUnited States and is now US Food and Drug Administration

(FDA)-approved is fecal elastase This is a stool

determina-tion from an aliquot of stool, not a quantitative collecdetermina-tion,and is measured by a radioimmunoassay of human fecalelastase Normal values are >200 µg/g of stool A value <200

µg/g but >100 µg/g usually reflects mild to moderate chronicpancreatitis A value <100 µg/g is severe PI and correlateswell with an abnormal fecal fat excretion or steatorrhea.These function tests reflect severe damage and often con-firm the abnormalities found on computed tomography(CT) or magnetic resonance imaging (MRI) In patients with

small duct disease who present largely just with pain and do

not have maldigestion, those function tests just mentionedare often normal Indeed in up to 40% of patients with smallduct chronic pancreatitis, radiographic tests including CT,MRI, magnetic resonance cholangiography (MRCP) andendoscopic retrograde cholangiography (ERCP) may be nor-mal It is those patients with small duct disease that haveradiographic negative findings that are the real challenge,particularly if the secretin hormone stimulation test is notavailable Patients suspected of having small duct chronicpancreatitis should be referred to a center that is proficient

in performing these tests When compared against any ographic test or function test as listed in Table 139-2, the

radi-secretin hormone stimulation test has consistently been found

to be more sensitive and more specific The secretin test hasbeen evaluated against histology in over 100 patients fromJapan In this study, the bicarbonate concentration of pan-creatic secretion was the most accurate parameter; ERCPwas 60% as accurate as the bicarbonate concentration(Hayakawa et al, 1992)

*Editor’s Note: Interestingly, two mutations in NOD 2/CARD 16

gene increase the risk of Crohn’s disease 20- to 40-fold.

TABLE 139-2 Diagnostic Tests for Chronic Pancreatitis

Until recently, the world supply of biologic porcine secretin

had been depleted because of the unproven use of this agent

to decrease the symptoms of autism Synthetic porcine

secretin and synthetic human secretin have been evaluated

against the biologic porcine secretin and both synthetic

preparations are equal to the biologic preparation when

evaluated against each other in normal subjects and patients

with proven chronic pancreatitis (Somogyi et al, 2003)

Synthetic porcine secretin is FDA-approved and the

application to approve synthetic human secretin for

clin-ical use is pending

EUS

In recent years, EUS has emerged as a forerunner among

radiographic tests evaluating pancreatic structure The use

of EUS overcomes two disadvantages that occur with

transabdominal ultrasonography, including the ability to

provide high resolution and not to be obscured by

over-lying bowel gas At our medical center, we have had

exten-sive experience with EUS and EUS has replaced ERCP as

a diagnostic test Whereas ERCP can image only the

pan-creatic ductal system, EUS can evaluate both the

pancre-atic ducts and the parenchyma We are continuing to

evaluate EUS and secretin against each other in patients

with suspected small duct disease who present with

chronic abdominal pain and negative radiographic

test-ing A preliminary report from our laboratory indicates

that EUS was only 57% as sensitive and 64% as specific as

a secretin test in patients with unexplained abdominal

pain who ultimately where shown to have small duct

chronic pancreatitis (Chowdhury et al, 2001) More

stud-ies have to be carried out comparing EUS to the secretin

hormone stimulation test in patients suspected of

hav-ing early chronic pancreatitis of the small duct variety

In those patients with a normal secretin test and EUS

evi-dence of changes of chronic pancreatitis, it is not clear

whether the EUS is more sensitive in some patients for

early changes or if it is truly over-diagnosing chronic

pan-creatitis (Draganov and Toskes, 2002) This is particularly

problematic because similar changes on EUS have been

found in an appreciable number of normal control

sub-jects and in people with nonulcerative dyspepsia Exactly

what the criteria should be for making a firm diagnosis of

chronic pancreatitis is still under discussion EUS is an

important modality in the diagnosis and management of

patients with chronic pancreatitis particularly if fine

nee-dle aspiration (FNA) is carried out Cystic lesions within

the pancreas, which may either be due to chronic

pancre-atitis or due to neoplastic causes, can be sorted out quite

well with EUS There is a separate chapter on EUS and

FNA (see Chapter 5, “Endoscopic Ultrasonography and

Fine-Needle Aspiration”)

Management

The cornerstone in medical therapy is the use of potentpancreatic enzyme preparations whether the clinician ismanaging maldigestion or abdominal pain Table 139-3lists some of the more commonly used pancreatic enzymeformulations in the United States They are divided intoconventional or nonenteric-coated preparations andenteric-coated preparations For the treatment of maldiges-tion, one to two capsules of an enteric-coated pancreaticenzyme preparation with meals is usually quite effective.With the use of these preparations, patients usually ceasehaving diarrhea, gain weight, and have an improved qual-ity of life Indeed if one measures in a quantitative fash-ion fecal fat excretion under a controlled high fat diet, itwill be shown that most patients treated with pancreaticenzymes do not completely correct their steatorrhea Isthere any danger in not correcting steatorrhea? This issuehas not been very clear but recent observations from sev-eral laboratories, including our own, have indicated thatpatients with pancreatic steatorrhea may have bonechanges of osteopenia and osteoporosis despite being ongood enzyme treatment and being asymptomatic It is alsointeresting that patients with small duct pancreatitis andpain but no steatorrhea have had similar bone changes Thenumbers of patients in these studies is small but work is inprogress to see whether or not this is of significant import

to patients with chronic pancreatitis Most experts have notrecommended that patients with chronic pancreatitis and

PI take supplements of calcium and vitamin D because itwas thought that the normal small intestine could com-pensate for the loss of calcium Indeed, our own observa-tions are such that when one calculates how much calcium

is lost per gram of fat lost, the presently allowed levels ofsupplementation would fall far short of correcting thisproblem in such patients

Pain

In respect to managing pain in patients with chronic creatitis, the pathogenesis is multifactorial and, thus, onetherapy is not apt to help all patients in a similar fashion

pan-TABLE 139-3 Pancreatic Enzyme Preparations

three times lipase content

from the pain caused by dysmotility There is a separatechapter on gastroparesis (see Chapter 31, “Gastroparesis”).

Coexistent Motility Disturbances

This interaction of motility disturbances with chronic creatitis takes on an even more complex connection

pan-because our laboratory has recently reported that theprevalence of gastroparesis is significantly increased inpatients with chronic abdominal pain and small ductchronic pancreatitis In 56 patients with small duct chronicpancreatitis documented by the secretin test, 25 of the 56(44%) had gastroparesis The etiology of this delay in gas-tric emptying is unclear, but it may be related to the highlevels of CCK that such patients with small duct chronicpancreatitis demonstrate In addition, many of thesepatients are receiving narcotics and that may lead to gas-troparesis Our laboratory has been quite successful in

using nonenteric-coated pancreatic enzymes to treat the pain

in patients with small duct chronic pancreatitis However, if

the stomach is not emptying properly, those enzymes willnot be delivered into the proximal small bowel where feed-back inhibition of pancreatic secretion is operative Thedocumentation of gastroparesis allows the administration

of prokinetics with the pancreatic enzymes to allow this

feed-back process to be restored Because motility disturbancesare quite frequent in patients with chronic pancreatitis, it

is strongly recommended that potent narcotic analgesics

be avoided if possible and anodynes, such as nonsteroidal

anti-inflammatory drugs, propoxyphene, or tramadol, beused These latter three medications do not slow down gas-tric emptying Despite all these efforts, narcotic addiction

is quite common in patients with chronic pancreatitis andconstant pain (see Chapter 41)

Feedback Control of Exocrine Secretion

Nonenteric Coated Pancreatic Enzyme Perspective

The basis for the use of nonenteric-coated pancreaticenzyme preparations to decrease abdominal pain inpatients with chronic pancreatitis is based on the concept

of feedback control of pancreatic exocrine secretion

(Isaksson and Ihse, 1983; Slaff et al, 1984) coated pancreatic enzyme preparations appear to inhibit

Nonenteric-pancreatic secretion through a negative feedback nism involving intraduodenal serine proteases in theexocrine pancreas These serine proteases modulate pan-creatic secretion by regulating CCK release Becausepatients with chronic pancreatitis often have decreasedintraduodenal protease activity, they may not be capable

mecha-of inactivating the CCK-releasing peptide that exists in theproximal small bowel and is largely responsible for stim-ulating CCK release In these patients, it can be demon-strated that there are high levels of CCK in the blood This

Chronic Pancreatitis / 799

The following three pathophysiologic mechanisms may

explain the pain in chronic pancreatitis: (1) acute pancreatic

inflammation, (2) increased intrapancreatic pressure, and

(3) alterations in pancreatic nerves Pancreatic enzyme

ther-apy, CCK antagonists and octreotide decrease pancreatic

secretion thus decreasing intrapancreatic pressure in both

large and small ducts, as well as the pancreatic gland itself

Endoscopic procedures attempt to decrease the pain by

decreasing pressure through improved drainage There is a

separate chapter on endoscopic treatment (see Chapter 138)

Surgery affords decompression of the major pancreatic duct

and is successful in some patients There is also a chapter on

surgical treatment (see Chapter 137, “Chronic Pancreatitis:

Surgical Considerations”) Celiac plexus block and

thoras-copic splanchnicectomy interrupt neural transmission of

pain signals There is also a chapter on chronic pain

man-agement (see Chapter 41, “Chronic Abdominal Pain”)

It must be appreciated that often the diagnosis of

chronic pancreatitis in a given patient is based on very

lit-tle evidence Mild elevations of pancreatic enzymes in the

blood of patients with chronic abdominal pain usually do

not mean chronic pancreatitis The clinician must

elimi-nate other causes of abdominal pain because the pain of

chronic pancreatitis is often nonspecific and can mimic

many other kinds of pain In particular, it is now being

appreciated that dysmotility syndromes both in the form

of gastroparesis and small bowel dysmotility can present

with abdominal pain indistinguishable from that of

chronic pancreatitis Patients with these dysmotility

syn-dromes may even manifest a mild elevation in serum

amy-lase and/or lipase There are separate chapters on nonulcer

dyspepsia (see Chapter 30, “The Management of Nonulcer

Dyspepsia”), chronic abdominal pain (see Chapter 41) and

intestinal pseudo-obstruction (see Chapter 63, “Chronic

Intestinal Pseudo-Obstruction”)

A recent study at our medical center presented

prelim-inary data on 74 patients referred to our pancreatitis clinic

who were suspected by the referring gastroenterologist and

surgeons as having radiographic negative chronic

pancre-atitis or small duct chronic pancrepancre-atitis (Chowdhury et al,

2001; Chowdhury et al, 2003) The examination of these

patients with unexplained pain of presumed pancreatic

origin revealed that 40% actually had chronic

pancreati-tis diagnosed by the secretin stimulation test and did well

with pancreatic enzyme treatments Fifty percent had

dys-motility most commonly of the stomach, and 10% had no

cause that could be found Prokinetic therapy in the form

of erythromycin ethyl succinate suspension given orally in

a dose of 100 mg 4 times a day, if the delayed gastric

emp-tying was mild to moderate, and erythromycin 200 mg

intravenously 4 times a day, if the gastroparesis was very

severe, decreased the abdominal pain in these dysmotility

patients Even retrospectively, the pain of chronic

pancre-atitis could not be distinguished in any accurate manner

is a proximal small intestine phenomenon and the

pan-creatic proteases must be delivered to the upper small

intes-tine This can be done consistently only by the

administration of nonenteric-coated pancreatic enzyme

preparations Such nonenteric-coated enzyme preparations

are susceptible to acid degradation as they pass through the

stomach; therefore, it is recommended that a proton pump

inhibitor (PPI) be given along with the pancreatic enzymes.

We have found that the most consistent preparation that

affords control feedback inhibition and relieves

abdomi-nal pain is the preparation of Viokase-16 This preparation

should be given orally in a dose of four tablets, four times

a day, along with the acid suppressive agent These enzyme

preparations are remarkably devoid of side effects The

remarkable side effect of colonic strictures noted in patients

with CF receiving very large doses of enteric-coated

pan-creatic enzymes (see Chapter 140, “Cystic Fibrosis and

other Hereditary Diseases of the Pancreas”) does not occur

in adult patients with chronic pancreatitis

Six randomized trials have evaluated the effectiveness

of pancreatic enzymes in the reduction of chronic

pancre-atitis pain (Warshaw et al, 1998) Of these, two trials using

nonenteric-coated enzymes were effective in reducing pain,

whereas the four trials using the enteric-coated

prepara-tions showed no statistical improvement in pain relief In

patients with big duct disease, there was, at best, a 25%

response rate in decreasing abdominal pain, whereas the

response rate was approximately 70% in those with small

duct disease A meta-analysis of these trials indicates that

pancreatic enzyme therapy did not decrease abdominal

pain in patients with chronic pancreatic (Brown et al,

1997) It should be pointed out that this meta-analysis has

been sharply criticized for lumping together patients

receiving nonenteric-coated and enteric-coated enzymes

Abundant information now exists both from randomized

trials and extensive clinical experience that

nonenteric-coated enzyme preparations are preferable for relief of the

abdominal pain in such patients Table 139-4 contrasts the

nonenteric and the enteric-coated enzymes in this

feed-back control process

Thus if one wants to optimize pancreatic enzyme

ther-apy in treating the pain of chronic pancreatitis, the

appro-priate enzyme preparation must be employed in a suitable

patient A nonenteric enzyme preparation should be given

along with a PPI to a patient with small duct chronic creatitis who does not have steatorrhea Great success will

pan-not be achieved when a patient with large duct disease andsteatorrhea is treated with an enteric-coated preparation

Duration of Enzyme Therapy

Our extensive experience indicates that those patients whoare receiving pancreatic enzyme therapy for pain and whoattain appreciable pain relief should be evaluated for dis-continuation of enzyme therapy if they have been pain-free for at least 6 months In our experience, about half ofthe patients no longer have pain when enzymes are dis-continued, whereas the other half experience a relapse andmust continue enzyme therapy indefinitely We have notyet been able to identify those markers that will allow us topredict in which group a given patient will fall

Octreotide

In patients with big duct disease, pancreatic enzyme apy is not very effective and may be beneficial in only 25%

ther-of patients Following the principle feedback inhibition,

it was proposed that octreotide, an analog of somatostatin,

might be effective in controlling pain Octreotide markedlyinhibits pancreatic secretion and significantly lowers CCKlevels Several small short-term studies result in variablefindings with regard to pain control with this agent Resultsfrom a multicenter, double-blind, placebo-controlled, dose-ranging pilot study suggests that a dose of 200 µg admin-istered subcutaneously 3 times a day was effective In thatstudy of over 100 patients, 65% of patients showed adecrease in abdominal pain with that dose of octreotideand a decrease in anodyne (pain relief medication) usage.Not all patients respond to this agent, but it appears thatthose that respond are those that demonstrate at least a50% reduction in blood CCK levels Studies are now beinglaunched from our laboratory that will be looking at a longacting form of octreotide that can be given once every 28days in the dose equivalent to the 200 µg 3 times a day(Draganov and Toskes, 2002) Thus in patients with smallduct chronic pancreatitis, the treatment of choice should

be nonenteric-coated pancreatic enzymes with a PPI.Treatment should be started and carried out for 4 to 6weeks before an evaluation of success or failure has beenmade If the patient fits all the criteria of small duct pan-creatitis and is not benefiting from the enzyme approachdescribed above, then a gastric emptying study should bedone and, if there is gastroparesis, a prokinetic should beadded to the plan

CCK Receptor Antagonists

A recent multicenter, dose-response controlled trial wasconducted in Japan to evaluate the efficacy of the CCK

receptor antagonist, loxiglumide, in patients with

abdom-TABLE 139-4 Pancreatic Enzyme Preparations and

Feedback Control of Pancreatic Secretion

CCK = cholecystokinin.

inal pain associated with chronic pancreatitis (Shiratori et

al, 2002) Two hundred and seven patients were randomly

assigned to several dosages of loxiglumide for 4 weeks or a

placebo The overall clinical improvement was 46 to 58%

in the loxiglumide categories versus 34% in the placebo

group In another study, a patient with abdominal pain

asso-ciated with chronic pancreatitis had elevated blood CCK

levels and analgesia produced by morphine appeared to be

reduced The patient was treated with proglumide, a

non-specific CCK receptor antagonist, and analgesia improved

markedly The patient derived consistent analgesia from the

use of proglumide alone It was theorized that

proglumide-induced analgesia would be reduced in chronic

pancreati-tis patients who have elevated CCK levels

Surgical Therapy

In patients with big duct pancreatitis and significant

abdominal pain, the treatment of choice is surgical ductal

decompression of the main pancreatic duct Usually in such

patients, 80% receive relief immediately, but if the patients

are followed for 3 years, there will be a decrease in the

num-ber of patients who have pain relief to the extent that 30 to

40% of patients will be free of pain (Howell et al, 2001)

There is a chapter on surgical therapy (see Chapter 137)

Endoscopic Therapy

In regard to endoscopic treatment, pancreatic stents have

been used in patients with chronic pancreatitis to reduce

intraductal pressure Such interventions are quite

appro-priate if there is a dominant stricture or a stone within the

duct There is no good evidence that a diffusely dilated

pan-creatic duct without any dominant stricture will respond

to stenting The possible benefits of pancreatic ductal

stent-ing must be balanced against the risk of further damage

induced by the stent to the pancreatic duct and the

pan-creatic parenchyma There is a chapter on endoscopic

ther-apy (see Chapter 138, “Pancreatitis: Endoscopic Therther-apy”)

Celiac Plexus Block

Celiac plexus block has had mixed results and, in our

opin-ion, is at best, a temporizing measure EUS-guided celiac

plexus block has been used in patients with both pancreatic

cancer and chronic pancreatitis When EUS-guided versus

CT-guided blocks are looked at in a prospective,

random-ized fashion, the EUS-guided celiac block seems to provide

more pain relief than the CT-guided block In patients

treated with ultrasonography-guided celiac plexus block,

a significant improvement in pain score occurred in 55%

of patients The benefit persisted beyond 12 weeks in only

26% of patients, and it appears that was not very effective

in patients under 45 years of age or those who have had

previous surgery (Gress et al, 2001) It appears that this

pro-cedure is well tolerated with temporary results that arequite impressive but its role at this time should probably

be limited to treating flares of chronic pain in patients withotherwise limited options This is discussed in the chap-ter on chronic abdominal pain (see Chapter 41)

ConclusionThe management of the abdominal pain in patients withchronic pancreatitis continues to pose significant chal-lenges to gastroenterologists and surgeons The subjectivenature of the pain, the heterogeneity of the patients, andthe poor understanding of the pathophysiology are poten-tial pitfalls This situation is exacerbated by concurrentalcohol abuse in some patients and narcotic dependence

in many, and it is further compounded by the lack of dictors for the response to pain Patients with chronicabdominal pain are quite common and often present totheir primary care physician for examination and man-agement If the cause of abdominal pain is not obvious byradiographic examination, a gastroenterologist should beconsulted for further diagnostic testing It is very impor-tant that a proper diagnosis be made and that the patho-physiology of the pain be considered rather than justtreating the symptoms What should not be done is anexpedient referral to a pain clinic without proper diagno-sis A clinical diagnosis of chronic pancreatitis based onscant evidence leads to the administration of narcotics for

pre-an indefinite period by such pain clinics With new ness that coexisting motility problems are common in suchpatients and the observation that narcotics commonly used

aware-by physicians that direct pain clinics make such motilitypatients worse, greater efforts should be made to probe thepathogenesis of the abdominal pain

Chronic pancreatitis should seriously be considered inall patients with unexplained abdominal pain It is againimportant to stress that chronic pancreatitis is not one dis-ease and all patients with chronic pancreatitis cannot betreated the same The importance of small duct versus largeduct disease must be emphasized If radiographic tests donot provide a diagnosis of chronic pancreatitis but the clin-ical suspicion is still high, then more sophisticated testingsuch as a direct hormone stimulation test or EUS should

be sought A nonenteric-coated pancreatic enzyme ration along with a PPI is the treatment of choice for thatsubset of patients who have chronic pain and small ductdisease Octreotide is increasingly being used for abdom-inal pain that is unresponsive to pancreatic enzyme ther-apy If medical therapy fails, an expertise must sought toattempt possible endoscopic and surgical management.Clinical trials with very high doses of pancreatic proteases,long acting forms of octreotide, and CCK antagonists inselected patients are underway

prepa-Chronic Pancreatitis / 801

Supplemental Reading

Brown A, Hughes M, Tenner S, et al Does pancreatic enzyme

supplementation reduce pain in patients with chronic

pancreatitis: a meta-analysis? Am J Gastroenterol 1997;92:2032–5.

Chey WY Neurohormonal control of the exocrine pancreas Curr

Opin Gastroenterol 1997;13:375–80.

Chowdhury R, Bhutani M, Forsmark C, et al The association of

gastroparesis with chronic pancreatitis Gastroenterology 2001;

20:A648.

Chowdhury R, Bhutani M, Mishra G, et al Comparative analysis

of pancreatic function testing versus morphologic assessment

by EUS for evaluation of chronic unexplained abdominal

pain Gastroenterology 2001;120:A647.

Chowdhury RS, Forsmark CE, Davis RH, et al Increased

prevalence of gastroparesis in small duct chronic pancreatitis.

Gress F, Schmitt C, Sherman S, et al Endoscopic

ultrasound-guided celiac plexus block for managing abdominal pain

associated with chronic pancreatitis: a prospective

single-center experience Am J Gastroenterol 2001;96:409–16.

Hayakawa T, Kondo T, Shibata T, et al Relation between pancreatic exocrine function and histological changes in chronic pancreatitis.

Shiratori K, Takeuchi T, Satake K, et al Clinical evaluation of oral administration of a cholecystokinin-A receptor antagonist (loxiglumide) to patients with acute painful attacks of chronic pancreatitis: a multi-center, dose-response study in Japan Pancreas 2002;25:E1–5.

Somogyi L, Ross AS, Cintron M, Toskes P Comparison of biologic porcine secretin, synthetic porcine secretin, and synthetic human secretin in pancreatic function testing Pancreas 2003;27:230–4 Slaff J, Jacobson D, Tillman CR, et al Protease-specific suppression

of pancreatic enzyme secretion Gastroenterology 1984;87:44–52 Warshaw AL, Banks PA, Fernandez-delCastillo C AGA technical review: treatment of pain in chronic pancreatitis Gastroenterology 1998;115:765–76.

804 / Advanced Therapy in Gastroenterology and Liver Disease

Peptic Ulcer Disease

The majority of patients with CF with pancreatic

insuffi-ciency (PI) have greatly reduced pancreatic production of

bicarbonate and consequent inability to raise the pH within

the duodenum This lower pH and the use of

anti-inflam-matory medication in some patients, such as ibuprofen and

prednisone, predispose patients to acid-peptic injury in the

duodenum This is another cause of symptoms of

abdom-inal pain, and may be difficult to diagnose without

endo-scopic investigation Assessment of the chest disease by the

attending respirologist or an anesthetist is warranted prior

to undertaking endoscopy Again, as with GERD, the

treat-ment is chronic acid suppression medication

Malabsorption and Maldigestion

The pancreatic disease, with severe fibrosis and cystic

changes on pathologic examination (Figure 140-1), is

pre-sent in the severe mutations, and accounts for the PI of these

patients CFTR in the pancreas is localized to the luminal

surface of ductular cells, and malfunction of this channel,

which is known to transport both chloride and

bicarbon-ate ions, leads to acidic and viscous secretions blocking

pan-creatic exocrine outflow With screening for panpan-creatic

function in early life, about 40% of infants have

pancre-atic sufficiency (PS) (defined as < 7% loss of ingested fat on

a 72-hour fecal fat collection), but by the age of about 5

years, this has declined to approximately 15% of cases This

screening for pancreatic function is not an easy task, as the

methods of checking for PS are difficult or invasive The

standard clinical test is a 72-hour fecal fat measure (Van de

Kamer et al, 1949) Loss of any stool during collection would

favour a more normal report of fecal fat loss In addition to

stool collection, a diet record of weighed food ingested

should be obtained for 5 days, with the stool collection

occurring during the last 3 days The average daily fat lost

in the stool is then assessed as a percent of the average dailyfat ingested The normal value is fecal loss of < 7 % ofingested fat Patients with PI will have a loss of fat usually

in the range of 20 to 50% This testing should be performed

on all patients at the time of diagnosis to assess pancreaticfunction It should also be repeated on patients who are ini-tially assessed as having PS when there is a change in GIsymptoms or any faltering of growth

Other testing for pancreatic function includes the use of

a monoclonal antibody based enzyme-linked say test for human pancreatic elastase 1 in the stool Analiquot from a 24-hour stool collection is assayed Humanelastase 1 is not present in porcine pancrease supplementalenzymes, so this test is convenient to perform while thepatient is receiving pancreatic enzyme supplements Thistest is both sensitive (96%) and specific (100%) for CFpatients with PI when compared with healthy subjects andthose with nonpancreatic diseases (Gullo et al, 1997; Soldan

immunoas-et al, 1997) It is also somewhat more acceptable than a hour fecal fat for those asked to perform the test This testwill sort patients into those that have PS and those who have

72-PI, and can be used to judge the requirement for pancreaticenzymes More recently it has also been used to track thedecline in exocrine pancreatic function in PS CF patients(Walkowiak et al, 2003) It cannot be used to test for the effi-cacy of enzyme therapy, or to monitor the effect of changes

or additions to therapy, such as the use of acid suppressants

A 72-hour fecal fat is needed for this

A further test for pancreatic function, the pancreaticstimulation test, involves insertion of a nasoduodenal tubeand the administration of secretagogues (secretin/chole-cystokinin) to stimulate the flow of pancreatic juice Thecollected secretions can then be assessed for pH and enzymecontent Technical challenges make these tests very demand-ing, and patients with PS have a significant chance of beingmisclassified as having PI (Schibli et al, 2003); therefore, this

is not usually performed in the clinical setting, but is veryuseful for research clinical investigations Endoscopic aspi-ration of pancreatic fluid following secretin stimulation isnot recommended as it has even greater potential for mis-classification of pancreatic functional status

NUTRITION

The majority of CF patients have PI, and consequent orrhea and azotorrhea, and are predisposed to gross mal-nutrition We know that the growth failure of children andyouth with CF is not intrinsic to the disease, but remedia-ble with provision of sufficient energy and protein alongwith supplemental pancreatic enzymes as needed There isalso evidence that improved nutrition helps stabilize lungfunction, improves quality of life, and may prolong life.Estimates of energy requirements for patients with CF areclearly higher than normal and should be based on weightgain and body fat stores The quality of the diet can fit with

steat-FIGURE 140-1 Pathologic examination of pancreatic disease.

normal recommendations for nutrient mix, recognizing

the importance of fat as the most energy dense nutrient,

and the one that makes our food most palatable Formerly

recommended restrictions on fat intake to improve GI

dis-comfort and other symptoms should be entirely abandoned.

Pancreatic replacement enzymes on the market at present

are derived from pig pancreas (eg, trade names Pancrease,

Cotazym, Creon, Ultrase), although newer crystalline

syn-thetic enzymes are in trials The appropriate use of enzymes

has been carefully examined in the last decade since the

first appearance of fibrosing colonopathy The Consensus

Statement (Borowitz et al, 2002,) of the Cystic Fibrosis

Foundation recommends the following dosing guidelines:

Infants: 2,000 to 4,000 Units lipase per 120 mL

for-mula or breast feed

Children < 4 years: 1,000 to 2,500 Units lipase per kg

Enzyme dose < 10,000 Units lipase per kg per day

Beyond these recommendations, lipase dosing based on

grams of fat in the diet is a rational approach Standards of

care support a maximum dose of 4,000 U lipase per gram

fat per day (Fitzsimmons et al, 1998) This cut off keeps

doses of lipase well below those reported with colonic

stric-tures The differential diagnosis of patients with ongoing

abdominal complications while taking the maximum

enzyme dose is listed in Table 140-2

As most of the enzymes in clinical use are enteric coated

to protect them from stomach acid, use of an H2RA or PPI

to raise duodenal pH can enhance enzyme function The best

test of improved absorption is a 72-hour fecal fat measure

Monitoring success in treating malabsorption and

mal-nutrition is done in children by the careful assessment of

growth throughout childhood and adolescence For adults,

maintenance of a normal and stable body mass index

(BMI) should be sought We want our young patients to

achieve normal growth, not only when compared with the

National Center for Health Statistics (NCHS)/Centers for

Disease Control and Prevention (CDC) growth curves(CDC Web site, <http://www.cdc.gov/growthcharts>), butwhen assessed by growth potential as estimated by mid-parental height Diligence should be taken in obtainingaccurate heights with a wall-mounted stadiometer, andweights with minimal standard clothing, and plotting thevalues on the CDC growth charts, including values for headcircumference and weight for height or BMI As well as thestandard measures, the Consensus Report on Nutrition(Borowitz et al, 2002) recommends mid-arm circumfer-ence and triceps skinfold be measured annually as a marker

of body composition Special vigilance on growth andnutritional status should be given at the time of diagnosis,during infancy, and during the pubertal growth spurt Weare much closer now to achieving normal growth of ourpatients when comparing height and weight to populationstandards, but there is room still for improvement.* Along with the loss of macronutrients, pancreatic insuf-ficient patients are at risk for deficiencies in fat-soluble vit-amins (A, D, E, and K), and essential fatty acids, and forthe consequences of these deficiencies particularly on bonemineral accretion, neurological function, and membraneintegrity Patients with PI are routinely supplemented with

fat-soluble vitamins at between one and two times the mal Much work on essential fatty acid levels (linoleic acid

nor-[18:2n-6] and alpha-linolenic acid [18:3n-3], and thelonger products, arachidonic acid [20:4n-6] and eicos-apentanoic acid [20:5n-3] and docosahexanoic acid [22:6n-3]) has been done Although recommendations for routinesupplementation have not been made, assurance of intake

of these fatty acids within the high fat diet is a prudent ommendation The quality of supplemental dietary fatshould be assessed by the dietitian to minimize saturatedand trans fatty acids As well, an emphasis on mediumchain triglycerides as a source of energy is unnecessary

rec-Routine annual monitoring of fat-soluble vitamin levels amin A, 25-OH-D,α-tocopheral and prothrombin time) is

(vit-currently recommended Other micronutrients includingiron, calcium and zinc, although clearly essential, have no

specific recommendations, except for a yearly hemoglobin and hematocrit as surrogates for iron nutriture, and ensur-

ing at least normal intake of the others

Nutrition support for those starting to fall below nel (percentile line) on the growth curves can be enhanced

chan-in a step-wise fashion as required, chan-includchan-ing oral mentation, nasogastric (NG) feeds for short term gain, andsurgically-placed enteral tube feeds for long term noctur-nal nutrition support (gastrostomy or jejunostomy tubes).There is some evidence that oral supplements merelyreplace food (Pencharz and Durie, 2000), but with somepatients weight gain occurs The energy provided by theseadditional supplements is often > 50% of the patients total

supple-TABLE 140-2 Differential Diagnoses of Patients With

Ongoing Abdominal Complaints While Taking the

Maximum Enzyme Dose

when caring for prepubescent adolescents with Crohn’s disease.

806 / Advanced Therapy in Gastroenterology and Liver Disease

energy intake However, with increased severity of

pul-monary disease, increased inflammation and anorexia,

patients are not able to ingest the required additional

energy, but do tolerate enteral infusions, usually provided

overnight This care is an additional significant burden for

families and requires a lot of support and encouragement

from the care team

Meconium Ileus

This is a neonatal disorder, which most often occurs in the

term infant, and is the presenting sign in about 15% of

chil-dren diagnosed with CF, usually those with PI Although no

specific CFTR mutations have been identified with

meco-nium ileus (MI), a modifier locus on human chromosome

19 has been determined (Zielenski et al, 1999) Suspicion

for MI may be raised with fetal ultrasounds (USs)

demon-strating dilated loops of bowel on the second or third

trimester scan Meconium plug syndrome, attributed to a

tight anus, should be ruled out In this disorder,

light-coloured gelatinous firm stool is present in the lower colon

and rectum The likelihood of MI is confirmed with a

bar-ium enema showing a microcolon About 50% of infants

have complications of the disorder with perforation,

meco-nium peritonitis, atresias, or volvulus A history of this should

be sought when seeing these patients later in life

For patients with uncomplicated ileus, a trial of

med-ical management may be successful in alleviating the

obstruction, but the infant should be at a center where

sur-gical intervention is available, if and when required As with

distal intestinal obstruction syndrome (DIOS) (see later in

chapter), a combination of acetyl-cysteine, polyethylene

gly-col (PEG) or water-soluble contrast can be used to clear the

obstruction For complications, a resection with primary

or secondary anastomosis may be performed For some

patients this results in loss of the ileocecal valve, loss of

absorptive surface in the ileum, and vitamin B12or bile salt

malabsorption This will then add to the PI of the child and

make malnutrition a greater risk Some patients with

atre-sia or volvulus may end up with a true short gut

DIOS

Previously known as meconium ileus equivalent, DIOS is

obstruction of the distal small bowel with inspissated stool

and mucous This occurs more frequently in the older child

adolescent or adult, and may present acutely, or have a

pro-drome of weeks or even months of nonspecific crampy

abdominal pain The patient presents with a partial or, less

commonly, complete distal small bowel obstruction with

crampy abdominal pain, vomiting, abdominal distension,

and decreased stooling Although the patients who have

DIOS are usually those with PI, fat per se is not likely the

major culprit, and decreasing the amount of fat in the diet

is never an appropriate measure There is some evidence

that it is the thick tenacious intestinal mucous that ates the intestinal block Some patients have problems withadherence to appropriate use of supplemental pancreaticenzymes, and encouraging better enzyme use does decreasethe bulkiness of the stool There is some evidence thatmotility of the bowel in CF is decreased, but there is no evi-dence the presently available prokinetic agents have anyrole In my experience, those who are chronically under-nourished, or maintain poor fluid intake, or experienceacute dehydration (eg, children who participate in a soc-cer tournament with inadequate drinking) put themselves

initi-at risk for these events, and may have recurrent episodes,

or a chronic rumbling history On physical examination,tenderness in the right lower quadrant is usually present,and there is a fullness or palpable mass in some patients

DIAGNOSIS

Conventionally, plain films of the abdomen, with supineand upright or lateral decubitus views, confirm air/fluidlevels, often with some dilation, in most of the small bowel

Frequently, there is a foamy appearance to the bowel

con-tents in the right lower quadrant The rectum is devoid ofair or stool, differentiating DIOS from constipation At our

center, this is followed by a real-time US of the bowel,

which is very helpful in examining intestinal contents, ing out intussusception, and observing bowel motility Theultrasonographer can often point out sluggish motility,

rul-or virtual absence of motility in the bowel segmentsclogged with “toothpaste” consistency stool The amount

of motility then guides the attending physician in the use

of therapy, whether a lavage solution from above, or mas from below, or a combination Physicians need to bemindful of the differential diagnosis (Table 140-3) and takesome care with the examination As stated above, simpleconstipation should be ruled out as this involves the rec-tosigmoid, or whole colon, and not the distal small bowel.Therapy includes use of products to clear the block,from above, if the obstruction is incomplete and there issome motility on the US examination, or from below, if theobstruction is fairly complete PEG 3350—electrolyte solu-tions (Golytely or others) can be given by lavage through

ene-TABLE 140-3 Differential Diagnoses for Patients Presenting with Abdominal Pain and Small Bowel Obstruction

Intussusception Adhesions (especially if patient had surgery for MI as newborn) Inflamed or ruptured appendix

Volvulus Crohn’s disease Fibrosing colonopathy

MI = meconium ileus.

an NG tube or other enteral feeding tube (G-tube or

J-tube) if available Most patients will not tolerate drinking

the quantity of these solutions required to clear DIOS The

volume to infuse varies from 5 to 15 mL/kg/hr to a

maxi-mum of 250 to 750 mL/hr Start slowly (50 to 100 mL/hr),

and increase every 10 to 15 minutes as tolerated until the

maximum rate tolerated is reached, and continue the lavage

until the effluent is clear and bile stained If the patient

becomes overly distended, or vomits, hold the lavage, and

clear from below, before advancing with the lavage

solu-tion again Although PEG solusolu-tions do not cause fluid

shifts, in my practice, patients have an intravenous

solu-tion running to provide maintenance fluids, and to correct

dehydration if present We allow patients to have small

quantities of clear fluids orally if desired, understanding

that some of the glucose in the fluids may then be

cotrans-ported with the electrolytes and thus absorbed; however,

this has not interfered with the GI clearance For patients

who are totally obstructed, who have little to no motility

on US, we would avoid lavage until some obstruction is

relieved Enemas from below can also be both diagnostic

and therapeutic Isotonic contrast will loosen the stool

block, and may also help rule out intussusception,

stric-tures, and non-DIOS lesions We also use acetyl-cyteine

(Mucomyst) as enemas This comes as a 20% solution, and

we dilute to 4%, and provide repeat enemas of 50 to 150

mL This helps to break up disulfide bonds in the mucous,

and although the literature is sparse (Hodson et al, 1976),

this does work

Patients with one episode of DIOS may go on to repeat

events, and chronic therapy should be instituted to prevent

recurrence if possible Discuss with patients the need to

titrate enzymes appropriately and to not miss doses This

may not be the whole answer, but will decrease stool bulk

Use of a regular stool softener, such as lactulose, milk of

magnesia, mineral oil, or PEG, can be given on a routine

basis Patient preference and adherence to care should

guide the choice of product Miralax may be an option

where available because it is far more palatable Usual doses

for all the above are those used to treat simple

constipa-tion In the past, chronic use of acetyl-cysteine has been

advocated, but there is only anecdotal evidence for its use

It can be given as a 4% solution (4 mL acetyl-cysteine with

16 mL beverage) mixed with a stronger beverage such as

cola, taken once or twice a day Mineral oil should not be

given at the same time as vitamin supplements

Intussusception

This is a relatively common occurrence in children with

CF, and may occur intermittently It is also an important

cause of intestinal obstruction in the adult with CF (Di

Sant’Agnese et al, 1979) As with DIOS, inspissated mucous

may play a role, in this case as the lead point for the

intus-suscipiens We have noted patients who have this

appear-ing and disappearappear-ing over a few minutes when examined

by US Clearly, it is only when the blood flow is mised, with swelling and bowel obstruction, that treatment

compro-is required Insufflation of the colon with air, or the use

of a contrast enema, may decompress the intussusception.Failing this, surgery will be required

Constipation

Patients with CF may have constipation, just as do otheradults and children History, physical examination includ-ing digital rectal examination, and a plain abdominal film

is of help in differentiating this from DIOS Patients with

PI at times decrease their enzyme dose with the mistakenthought that this will loosen the stool and treat the con-stipation Physicians need to ensure that patients are tak-ing appropriate doses of enzymes Usual measures ofincreasing dietary fluid and fiber should be taken, butattention needs to be given to ensure the energy content ofthe diet does not suffer with this decrease in energy den-sity The usual stool softeners, such as mineral oil, milk ofmagnesia, or PEG solutions, can be titrated to effect

Fibrosing Colonopathy

This disease was first described in CF patients from Britain

in 1994 in a report from Dr R Smyth About 2 years beforethe report, very high dose supplemental pancreatic enzymeproducts became available on the market, with lipase con-tent of 20,000 or 25,000 U/capsule Although this was done

to limit the number of capsules the patient had to take,these higher lipase capsules could more easily providepatients with exceptionally high lipase doses This was espe-cially true when patients made little change in the number

of capsules when switching from a lower dosage to a higherdosage This initial report described five children with CFwho developed strictures, chronic obstructive symptoms,and eventually underwent colectomy The pathologic

description of the surgical specimens was marked cosal fibrosis primarily affecting the proximal colon.

submu-Subsequent work indicates that before that final stage is

reached, the endoscopic appearance may be that of a few nonspecific ulcers Again the microscopic appearance of

mucosal biopsies shows nonspecific changes with fairlymarked eosinophilia, but no granulomas The epidemiol-ogy of the disease from a good case-control study in theUnited States points to the dose of pancreatic enzyme asthe causative factor The fibrosis is the final outcome forcolons that have been irreparably damaged by very highdose enzymes Based on observations in the rat, onehypothesis about the nature of the injury involves theincreased intestinal permeability in CF patients (Mack et

al, 1992), in addition to high dose pancreatic enzymes ing to enteropathy, fibrosis, and possibly hepatic injury(Lloyd-Still et al, 1998) Current recommendations limit

lead-not controlled, or those who have gastric varices that

can-not be banded, surgical decompression by transjugular

intrahepatic portosystemic shunt or surgical shunt such as

splenorenal shunt may be required Some patients who

continue to bleed despite these aggressive measures, may

need to be listed for transplant as the ultimate means of

decompressing the portal circulation

Many gastroenterologists will also think about going on

to secondary prevention of bleeding varices with

nonselec-tive β-blockers, such as propranolol, to decrease

splanch-nic pressure and cardiac output (Shashidhar et al, 1999) The

respirologist caring for the patient needs to be consulted

before starting this therapy to ensure it will not interfere with

other therapy, such as the use of bronchodilators for lung

disease Some centers would avoid the use ofβ-blockers

except in those cases where bleeding cannot be controlled

by banding, such as patients with gastric varices or portal

hypertensive gastropathy Should the drug need to be

stopped at some point, the risk for rebleeding appears to

return to baseline risk Primary prevention of variceal

bleed-ing in CF has not been well studied Gastorenterologists who

consider eitherβ-blockers or banding must take into account

the severity of the pulmonary disease, other medications the

patient may be on, and the risk for general anesthesia All

these may argue against treatment before the first bleed

Patients with complications of portal hypertension, such

as bleeding, ascites, and malnutrition, are candidates for

liver transplant, and referral should be made at an early

stage, taking into account the severity of the lung disease in

the patient We have shown that survival is good and

qual-ity of life is improved posttransplant (Mack et al, 1995)

Some patients do well with transplant of liver and lungs

Biliary Obstruction

The viscous biliary secretions may also lead to

extrahep-atic biliary obstruction, with gallbladder sludge and

cholelithiasis occurring more commonly than normal

There is also some work demonstrating extrinsic

com-pression of the common bile duct by the fibrotic pancreas

Standard imaging studies for these presentations should

be performed and endoscopic retrograde

cholangiopan-creatography (ERCP) is also a useful tool

Shwachman-Diamond Syndrome

Shwachman-Diamond syndrome (SDS) is an autosomal

recessive genetic disorder that is caused by mutations of

the SBDS gene (Boocock et al, 2003) Although the exact

function of the SBDS protein is currently unknown, the

most common clinical manifestations are characterized by

exocrine pancreatic dysfunction, bone marrow

dysfunc-tion, and skeletal abnormalities (Mack et al, 1996) Other

organs may be involved including the liver, kidneys, heart,

central nervous system, and teeth The difficulty in

diag-nosis lies in the variability of disease expression amongindividuals and the clinical symptoms can vary with age(Ginzberg et al, 2000)

DIAGNOSIS

There are several mutations in the SBDS gene discovered

to date and with the phenotypic variation among uals, gene testing may become a tool similar to its use in

individ-CF as this form of testing becomes more available By andlarge the diagnosis is made in infancy and is based on theconstellation of clinical features, along with blood tests andradiologic investigations Acinar and ductal exocrine pan-creatic dysfunction may be quantified by pancreatic stim-ulation tests but they are demanding and no standardmethodology has been established Serum testing for pan-creatic enzymes may be useful as a diagnostic tool in sus-pected patients because serum pancreatic isoamylaseremains low in SDS patients in contrast to the serumcationic trypsinogen that may increase with advancing age.Bone marrow dysfunction is characterized in peripheralblood counts by persistent or intermittent anemia,leukopenia, and/or thrombocytopenia (Dror et al, 2002).Radiologic investigations reveal abnormal development ingrowth plates and metaphyses (Makitie et al, 2004) andimaging studies show a small fatty pancreas

Management Issues

STATURE ANDSKELETALPHENOTYPES

It is not unusual for the patients with SDS to have shortstature throughout life The mean height and weight ofpatients is below the 5th percentile but after infancy growthvelocity normalizes and so longitudinal measurementsshow height and weight measurements paralleling butbelow the 5th percentile However, as in all manifestations

of SDS, there is variability and some SDS patients can haveheights above the 25th percentile Pancreatic enzymereplacement therapy to normalize digestion will not reversethe short stature No phenotype-genotype correlationshave been recognized for skeletal findings

Skeletal changes are present in all patients but any givenspecific abnormality is age-dependent Due to the effects

of metaphyseal chondrodysplasia of the femur, patients canhave persistent asymmetrical growth resulting in valgusdeformities of the head and neck of the femur and varusdeformities of the knees In addition, some patients canhave structural failure of metaphyseal bone of the femoralnecks giving rise to fractures and varus deformities.Osteopenia and compression fractures have been docu-mented in SDS patients Patients should have their bonemineral density monitored Serial radiographic determi-nations in patients with skeletal deformities or examina-tions of sites of bony pain (eg, spinal radiographs forvertebral compression fractures) should be performed

810 / Advanced Therapy in Gastroenterology and Liver Disease

Orthopedic interventions may be needed With exocrine

pancreatic involvement being a prominent feature of SDS

patients, vitamin D status should be monitored and

cor-rected by ensuring good calcium and vitamin D intake and

adequate pancreatic enzyme replacement therapy for

pan-creatic insufficient patients

EXOCRINEPANCREASFUNCTION

SDS is the second most common inherited cause of

exocrine PI after CF Unlike pancreatic disease in CF,

creatitis is not a feature of SDS and there is normal

pan-creatic fluid and bicarbonate secretion (Stormon and

Durie, 2002) Virtually all patients have some degree of

exocrine pancreatic dysfunction though the degree of

dys-function is variable Thus, quantitative tests of exocrine

pancreatic function remain abnormal but some patients

may not require pancreatic enzyme replacement therapy

for normal fat digestion (ie, they are pancreatic sufficient)

Furthermore, about 50% of patients may show

improve-ment in exocrine pancreatic function with advancing age

such that they become pancreatic sufficient and no longer

need pancreatic enzyme replacement therapy

Consequently, if a patient with this condition has not been

re-evaluated for some time to verify whether there is a

con-tinued need for pancreatic enzyme replacement therapy

then obtaining a serum trypsinogen would be useful If the

serum trypsinogen is in the intermediate or normal range

(> 6 µg/L) then performing a 72-hour fecal fat collection

to evaluate fat digestion is indicated One must

remem-ber that even though the serum trypsinogen and fecal fat

determinations may improve with age, the serum

pancre-atic isoamylase and possibly other pancrepancre-atic digestive

enzymes do not Consequently, there may continue to be

benefit from pancreatic enzyme replacement capsules as

they contain a mixture of enzymes Fat-soluble vitamin

deficiencies should be monitored and corrected with

sup-plementation

HEMATOLOGYDYSFUNCTION

All cellular lines of the bone marrow may be abnormal

The most common hematologic abnormality is

neutrope-nia and it is usually intermittent As well, anemia with low

reticulocytes and thrombocytopenia may be identified As

a result of bone marrow dysfunction SDS patients are at

risk of bleeding, developing severe infections, and

suffer-ing from periodontal disease In cases of life threatensuffer-ing

infection, granulocyte-colony stimulating factor (G-CSF)

may be required All three cell lines can be involved, and

patients with this complication are at a greater risk of

devel-oping severe aplasia, advanced myelodysplastic syndrome,

or acute myeloid leukemia SDS-related leukemia carries a

poor prognosis Ongoing consultation with an

hematolo-gist would be advisable to decide how often to perform

blood tests and bone marrow aspirations One suggestionhas been to recheck blood work every 4 months and per-form yearly bone marrow aspirations in SDS patients with-out complications and, for patients with severe cytopenia,

to have blood tests repeated every 1 to 3 months with bonemarrow aspirations performed every 3 to 6 months

The development of chronic pancreatitis is the end-result

of a process whereby recurrent acute pancreatitis eventsoccur because of increased susceptibility, triggering eventsand the development of a fibrotic and destructive response.Pancreatitis may develop as the result of both intra-acinar

or intraductal events Molecular techniques applied togroups of patients and families with a high prevalence ofpancreatitis are yielding information as to mechanismswhereby gene mutations increase susceptibility in thedevelopment of acute pancreatitis events In the pancreaticparenchymal cells an imbalance in the activation of pro-teases and their inhibition may be integral to the process.Small amounts of active trypsin are normally generatedfrom its inactive precursor, most notably cationic trypsino-gen Trypsin molecules are kept in check by proteaseinhibitors, such as serine protease inhibitor Kazal type 1(ie, SPINK1; also known as pancreatic secretory trypsininhibitor or PST1) and autolysis However, when 10 to 20%

of cationic trypsinogen (ie, protease, serine 1; PRSS1)becomes activated, the inhibitory SPINK1 mechanismbecomes overwhelmed and a cascade of events can followwith the end result of pancreatitis (Whitcomb, 2002; Witt,

2003; Naruse, 2003) Some mutations in PRSS1 (eg,

arginine-histidine substitution at residue 122 [R122H])result in increased autoactivation and yield a trypsin resis-tant to autolysis, whereas other mutations, such as theasparagine-isoleucine substitution at residue 29 (N29I),appear to have increased autoactivation only Mutations in

SPINK1 (eg, asparagines-serine substitution at residue

34-N34S) result in loss of the SPINK1 line of defense ing in more intracellular trypsin It is the acinar ductal cells

result-where expression of CFTR occurs CFTR mutations have

been classified by the CF genetic analysis into variousclasses based on their predicted molecular dysfunction, andthere are a number of proposed mechanisms whereby pan-creatitis can develop based on the fluid and electrolytealterations in the duct with the resultant effect of acinarinflammation (Freedman et al, 2000) Disparate clinicalconsequences might be predicted at different gene muta-

tion sites but interestingly any given mutation has a

spec-trum of phenotypic expression A number of mutations in

the above 3 genes have been described, including over 1,000

for CFTR and around 10 so far for PRSS1 and SPINK1

(http://archive.uwcm.ac.uk/uwcm/mg/hgmd0.html) The

spectrum of phenotypic expression between individuals

remains unexplained but may be as a result of

combina-tions of mutacombina-tions For instance, CF individuals

homozy-gous for the most common CFTR mutation (deletion of

the phenylalanine residue at position 508 [ie,∆F508]) have

evidence of pancreatitis early in life (fetal and neonatal)

and generally are pancreatic insufficient Individuals that

present with chronic pancreatitis rather than typical CF

manifestations may be compound heterozygotes for CFTR

(Cohn et al, 2002) As well, there may be synergism with

non-CFTR gene modifiers For instance, mutational

analy-sis of patients with chronic pancreatitis has yielded

indi-viduals trans-heterozygous for a CFTR mutation and a

mutation in SPINK1 (Audrezet et al, 2002; Noone et al,

2001) Thus, mutations in genes associated with

pancre-atic functioning can create an environment in the pancreas

that modifies responses to pancreatic insults and places

individuals at risk of developing pancreatitis Then, other

processes lead to the development of chronic pancreatitis

and tissue destruction

Patient Counseling and Monitoring

The clinical signs and symptoms of pancreatitis in

indi-viduals with genetic mutations such as R122H in PRSS1

are not different from those induced by other causes, such

as alcohol (Whitcomb et al, 2002) However, the presence

of hereditary mutations such as R122H in

PRSS1-hered-itary pancreatitis and CFTR in CF are significant risk

fac-tors for development of pancreatic cancer (Lowenfels et

al, 2000) The presence of pancreatitis-associated

muta-tions (eg,∆F508, R122H) themselves is not important in

the development of pancreatic cancer However, the

pancreatitis-associated genes result in an environment of

chronic inflammation that is postulated to increase the

penetrance of other germline mutations that promote

oncogenesis of the pancreas It would seem quite

reason-able to counsel patients to avoid alcohol intake because it

is a controllable environmental risk factor in the etiology

of pancreatitis Smoking appears to double the high risk

of pancreatic cancer and lower the mean age of

develop-ment of pancreatic cancer by 20 years for those with

hereditary pancreatitis Thus, avoidance of cigarette

smok-ing should be advocated because it is also a controllable

environmental factor associated with the development of

pancreatic cancer There is a separate chapter on smoking

and GI diseases (see Chapter 45, “Smoking and

Gastrointestinal Disease”) Because of the significant risk

in the development of pancreatic cancer, screening for it

(endoscopic US, helical CT, ERCP) should be offered to

those patients at the age of 40 years with known genemutations (Ulrich, 2001) There is a separate chapter oncysts and precancerous lesions of the pancreas (seeChapter 143, “Neoplastic Cysts and Other PrecancerousLesions of the Pancreas”) Other long term concernsrelated to the destruction of the pancreas include thedevelopment of PI in 20% of patients and the develop-ment of diabetes mellitus in 7.5% Assessment and appro-priate therapy for both should be part of the long termfollow-up of these patients (Sossenheimer et al, 1997)

GENESCREENING

Phenotype penetrance is not 100% related to genotype,and genetic testing can have significant psychological andpractical implications for the individual regarding lifestyle,work, and insurability Information gained from geneanalysis must be balanced with patient implications A con-

sensus conference recommended testing for PRSS1 R122H

and N29I for patients with unexplained recurrent acutepancreatitis, unexplained chronic pancreatitis, and a fam-ily history of chronic pancreatitis Additionally, for chil-dren with unexplained pancreatitis the consideration of

the above PRSS1 mutations and SPINK1 N34S mutation

can be justified (Ellis et al, 2001) Screening of a greaternumber of mutations has been also advocated since theconsensus conference report but still only half of chronicpancreatitis patients were found to have one of the cur-rently known mutations

Boocock GRB, Morrison JA, Popover M, et al Mutations in SBDS are associated with Shwachman-Diamond syndrome Nat Genet 2003;33:97–101.

Borowitz D, Baker RD, Stallings V Consensus report on nutrition for pediatric patients with cystic fibrosis J Pediatr Gastroenterol Nutr 2002;35:246–59.

Borowitz D, Grand RJ, Durie PR Use of pancreatic enzyme supplements for patients with cystic fibrosis in the context

of fibrosing colonopathy J Pediatr 1995;127:681–4 Cohn JA, Noone PG, Jowell PS Idiopathic pancreatitis related

to CFTR: complex inheritance and identification of a modifier

gene J Invest Med 2002;50:247–55S.

Di Sant’Agnese PA, David PB Cystic fibrosis in adults: 75 cases and a review of 232 cases in the literature Am J Med 1979; 66:121–32.

Dror Y, Freedman MH Shwachman-Diamond syndrome Br J Haematol 2002:118:701–13.

Ellis I, Lerch MM, Whitcomb DC, et al Genetic testing for

hereditar y pancreatitis—guidelines for indications,

counseling, consent and privacy issues Pancreatology

2001;1:405–15.

Fitzsimmons SC, Burkhart GA, Borowitz D, et al High-dose

pancreatic-enzyme supplements and fibrosing colonopathy

in children with cystic fibrosis N Engl J Med 1998;336:1283–9.

Fomon SJ, Ziegler EE, Thomas LN, et al Excretion of fat by

normal full-term infants fed various milks and formulas Am

J Clin Nutr 1970;23:1299–313.

Freedman SD, Blanco P, Shea JC, et al Mechanisms to explain

pancreatic dysfunction in cystic fibrosis Gastroenterol Clin

North Am 2000;84:657–64.

Ginzberg H, Shin J, Ellis L, et al Segregation analysis in

Shwachman-Diamond syndrome: evidence for recessive inheritance Am J

Hum Genet 2000;66:1413–6.

Gullo L, Graziano L, Babbini S, et al Faecal elastase 1 in children

with cystic fibrosis Eur J Pediatr 1997;156:770–2.

Hendriks HJE, van Kreel B, Forget PP J Pediatr Gastrenterol Nutr

2001;33:260–5.

Hodson ME, Mearns MB, Batten JC BMJ 1976;2:790–1.

Lenaerts C, Lapierre C, Patriquin H, et al Surveillance for cystic

fibrosis-associated hepatobiliary disease: early ultrasound

changes and predisposing factors J Pediatr 2003;143:343–50.

Lloyd-Still JD, Uhing MR, Arango V, et al The effect of intestinal

permeability on pancreatic enzyme-induced enteropathy in

the rat J Pediatr Gastroenterol Nutr 1998;26:489–95.

Lowenfels AB, Maisonneuve P, Whitcomb DC Risk factors for cancer

in hereditary pancreatitis Med Clin North Am 2000;84:565–72.

Mack DR, Flick JA, Durie PR, et al Correlation of intestinal

lactulose permeability with exocrine pancreatic dysfunction.

J Pediatr 1992;120:696–701.

Mack DR, Forstner GG, Wilschanski M, et al Shwachman

syndrome: exocrine pancreatic dysfunction and variable

phenotypic expression Gastroenterology 1996;111:1593–602.

Mack DR, Traystman MD, Colombo JL, et al Clinical denouement

and mutation analysis of patients with cystic fibrosis undergoing

liver transplantation for biliar y cir rhosis J Pediatr

1995;127:881–7.

Makitie O, Ellis L, Durie PR, et al Skeletal phenotype in patients

with Shwachman-Diamond syndrome and mutations in

SBDS Clin Genet 2004;65:101–12.

Naruse S Molecular pathophysiology of pancreatitis Intern Med

2003;42:288–9.

Neglia JP, Fitzsimmons SC, Maisonneuve P, et al The risk of

cancer among patients with cystic fibrosis N Engl J Med

1995;11:434–7.

Noone PG, Knowles MR ‘CFTR-opathies’: disease phenotypes

associated with cystic fibrosis transmembrane regulator

gene mutations Respir Res 2001;2:238–332.

Noone PG, Zhou Z, Silverman LM, et al Cystic fibrosis gene mutations and pancreatitis risk-relation to epithelial ion transport and trypsin inhibitor gene mutations Gastroenterology 2001;121:1310–9.

Pencharz PB, Durie PR Pathogenesis of malnutrition in cystic fibrosis, and its treatment Clin Nutr 2000;19:387–94 Schibli S, Corey M, Durie P The pancreatic stimulation test– factors that influence validity [abstract] J Pediatr Gastroenterol Nutr 2003;37:361–2.

Shashidhar H, Langhans N, Grand RJ Propranolol in prevention

of portal hypertensive hemorrhage in children: a pilot study.

J Pediatr Gastroenterol Nutr 1999;29:12–7.

Smyth RL, Van Velzen D, Smyth AR, et al Strictures of ascending colon in cystic fibrosis and high-strength pancreatic enzymes Lancet 1994;343:85–6.

Soldan W, Henker J, Sprossig C Sensitivity and specificity of quantitative determination of pancreatic elastase 1 in feces of children J Pediatr Gastroenterol Nutr 1997;24:53–5 Sossenheimer MJ, Aston CE, Preston RA, et al Clinical characteristics of hereditary pancreatitis in a large family, based

on high-risk haplotype Am J Gastroenterol 1997;92:1113–6 Stormon MO, Durie PR Pathophysiologic basis of exocrine pancreatic dysfunction in childhood J Pediatr Gastroenterol Nutr 2002;35:8–21.

The Cystic Fibrosis Genetic Analysis Consortium Population variation of common cystic fibrosis mutations Hum Mutat 1994;4:167–77.

Ulrich CD II Pancreatic cancer in hereditary pancreatitis— concensus guidelines for prevention, screening, and treatment Pancreatology 2001;1:416–22.

Van de Kamer JH, Huinunk HTB, Weyers HA Rapid method for the determination of fat in feces J Biol Chem 1949;177:347–55 Walkowiak J, Nousia-Arvanitakis S, Agguridaki C, et al Longitudinal follow-up of exocrine pancreatic function in pancreatic sufficient cystic fibrosis patients using the fecal elastase-1 test J Pediatr Gastroenterol Nutr 2003;36:474–8 Whitcomb DC, Pogue-Geile K Pancreatitis as a risk factor for pancreatic cancer Gastroenterol Clin North Am 2002;31: 663–78.

Whitcomb DC How to think about SPINK and pancreatitis Am

J Gastroenterol 2002;97:1085–8.

Witt H Chronic pancreatitis and cystic fibrosis Gut 2003;52:ii31–41 Zielenski J, Corey M, Rozmahel R, et al Detection of a cystic fibrosis modifier locus for meconium ileus on human chromosome 19q13 Nat Genet 1999;22:128–9.

812 / Advanced Therapy in Gastroenterology and Liver Disease

Patients may present with mass-related signs or symptoms

or conditions caused by excess hormones or vasoactivepeptides Any of the lesions may present with pain but typ-ically, patients with periampullary ACs present with pain-less obstructive jaundice and/or gastric outlet obstruction.All tumors that secrete hormones or vasoactive peptidesmay present with the signs and symptoms of the specifichormonal excess If not hormonally active, distal lesionstend to present with pain and or signs and symptoms ofwidespread metastases

Work-Up

The great majority of patients presenting with pancreatic

or periampullary neoplasms are studied with helical puted tomography (CT) scan, which anatomically placesthe lesion and gives information about morphology, such

com-as whether it is solid or cystic A multidetector, dual-phcom-ase

CT scan of the pancreas with three-dimensional structions is preferable This technique gives exquisite detailabout the possible involvement of the nearby vascularstructures, such as the portal vein, superior mesenteric vein(SMV), superior mesenteric artery (SMA), celiac axis, andhepatic artery Additionally, the liver and peritoneal cavitycan be screened for possible involvement at the same time.Because the chest also represents a potential site of metasta-tic spread, it should also be screened This may be accom-plished by either a chest CT or chest radiograph

recon-Patients who present with biliary duct obstruction can befurther worked up with biliary imaging These include bothinvasive and noninvasive techniques The most common

Incidence

Pancreatic and periampullary neoplasms are a diverse group

of tumors The great majority of these lesions are

pancre-atic cancers, which have an annual incidence of

approxi-mately 28,000 cases per year in the United States (Lillemoe

et al, 2000) Pancreatic cancer is the fifth leading cause of

cancer-related deaths (following lung, colon, breast, and

prostate), and is responsible for 5% of all cancer-related

deaths A more rare cause of pancreatic and periampullary

neoplasms are neuroendocrine tumors with a reported

inci-dence of 4 to 10 per million (Ahrendt and Demeure, 2001)

The incidence of pancreatic cystic neoplasms is not well

characterized, but with the more frequent use of

cross-sectional imaging, many more patients with asymptomatic

lesions are being discovered There is a separate chapter on

this topic (see Chapter 143, “Neoplastic Cysts and Other

Precancerous Lesions of the Pancreas”) The causes of

pan-creatic and periampullary neoplasms are listed in Table

141-1 and Table 141-2 For anatomic and technical reasons,

neoplasms of the pancreas can be divided into distal lesions

(potentially require resection of the neck, body, and/or tail)

and periampullary (potentially require

pancreaticoduo-denectomy) As can be seen by Table 141-1 and Table 141-2,

the neoplasms that comprise this group of diseases are quite

numerous and diverse Pancreatic adenocarcinoma (AC)

TABLE 141-1 Pancreatic Neoplasms

Acinar cell carcinoma

Giant cell carcinoma

Solid pseudopapillary tumor

Lymphoma

Cystic

Malignant potential

Intraductal papillary mucinous neoplasm

Mucinous cystic neoplasm

Usually behave in benign fashion

Serous cystic neoplasm

TABLE 141-2 Periampullary Neoplasms (Including Adenocarcinomas)

Pancreatic neoplasms (see Table 141-1) Distal bile duct neoplasms

Ampulla of Vater neoplasms Duodenal neoplasms

814 / Advanced Therapy in Gastroenterology and Liver Disease

technique is magnetic resonance

cholangiopancreatogra-phy This technique has the advantage of being

noninva-sive and can be accomplished without bacterial seeding

of the obstructed biliary system, and without the risk of

bleeding or pancreatitis This technique has the

disadvan-tages of being without the possibility of making a tissue

diagnosis or draining an obstructed biliary system The

common invasive techniques include endoscopic

retro-grade cholangiopancreatography and percutaneous

trans-hepatic cholangiography with or without drainage These

techniques have the converse advantages and disadvantages

just previously discussed Obstructive jaundice does not

need to be relieved prior to bringing a patient to the

oper-ating room if the patient is not suffering any septic or

nutri-tional consequences of the obstruction Several authors

have reported on the increased rate of infectious and

bleed-ing complications in patients who have undergone

manip-ulation of their biliary trees prior to definitive resection

In certain cases, endoscopic ultrasound (EUS) with or

without biopsy may be of benefit This technique can

espe-cially be helpful in smaller lesions not well characterized

by CT or magnetic resonance imaging This test should

only be used in situations where the outcome will effect

subsequent management For example, it is rare that EUS

and biopsy changes the decision to explore an elderly

patient with painless obstructive jaundice, a good quality

CT scan demonstrating a resectable solid mass in the head

of the pancreas, no evidence of metastatic disease, and who

is a good operative candidate Additionally, patients with

a good quality three-dimensional CT demonstrating

resectability might be best served with exploration even if

the EUS shows possible vessel involvement because of the

low but real false-positive rate of the test EUS and fine

nee-dle aspiration are discussed in a separate chapter (see

Chapter 5, “Endoscopic Ultrasonography and Fine-Needle

Aspiration”)

Patients who present with signs and symptoms of

hor-monal excess or who are suspected of having a

neuroen-docrine tumor may benefit from other specialized testing

The value of performing the various specific biochemical

and imaging tests are quite variable from patient to patient

and often depend on the degree of suspicion and possible

clinical consequences Each of the specific neuroendocrine

disorders that can affect the pancreas or the periampullary

region has specific hormones or peptides that can be

assayed for in the blood or urine Additionally, the

major-ity of pancreatic and periampullary neuroendocrine lesions

will be detectable with octreotide scan It is sometimes of

benefit to place patients with functional carcinoid lesions

on octreotide prior to resection to block the potential

sys-temic consequences of a sudden release of serotonin with

manipulation There are separate chapters on secretory

diarrhea (see Chapter 72,“Secretory Diarrhea”) and on

gas-trinoma (see Chapter 31, “Gastroparesis”)

Potentially Curative Treatment

ampullary neoplasms are usually accomplished with creaticoduodenectomy Patients are sometimes explored first

pan-with the intent of trying a transduodenal ampullectomyand/or bile duct exploration for small and superficial lesionsthought to have a high chance of being benign and/or to ruleout stone disease The decision to proceed with pancreati-coduodenectomy can be made at the time of the transduo-denal procedure, depending on the operative findings orfrozen sections

Once the decision has been made to proceed with a creaticoduodenectomy, exposure is accomplished either

pan-through a vertical midline or a bilateral subcostal incision.The first portion of this procedure is devoted to assessing theextent of disease and resectability There is debate as to thebenefits of staging laparoscopy versus open staging in antic-ipation of surgical resection or palliation At open explo-ration, the entire peritoneal cavity is assessed for the presence

of metastases not seen by preoperative imaging in studies.Tumor-bearing nodes within the resection zone do not con-traindicate resection because long-term survival is sometimesachieved with peripancreatic nodal involvement An exten-sive Kocher maneuver is performed by elevating the duode-num and head of the pancreas out of the retroperitoneumand into the midline, allowing the visualization of the SMA

at its origin at the aorta The porta hepatis is assessed bymobilizing the gallbladder out of its fossa and dissecting thecystic duct down to the junction of the common hepatic andcommon bile duct The hepatic artery is also assessed to deter-mine that it is free of tumor involvement

If the intraoperative assessment reveals localized diseasewithout tumor encroachment upon resection margins, theresection is performed in relative standard fashion If assess-ment reveals evidence of local tumor extension giving theearly impression of unresectability, the normal sequence forperforming the pancreaticoduodenectomy is modified sothat the easiest and safest portions of the resection are per-formed first, and the more difficult portions are performedlater In cases with localized disease without tumor encroach-ment upon resection margins, the distal common hepaticduct is divided close to the level of the cystic duct entry siteearly during the operation The gastroduodenal artery is nextidentified and divided For a pylorus-preserving pancreati-

coduodenectomy, the proximal gastrointestinal (GI) tract is

divided 2 to 3 cm distal to the pylorus with a linear stapling

device A plane is then formed between the neck of the

pan-creas and the underlying anterior surface of the portal vein

For a classic Whipple procedure, a 30 to 40% distal

gastrec-tomy is performed using a linear stapling device (Figure

141-1) The GI tract is divided distally at a point of mobile

jejunum, typically 20 cm distal to the ligament of Treitz The

proximal jejunum is then separated from its mesentery and

delivered dorsal to the superior mesenteric vessels from the

left to the right side The SMV caudal to the neck of the

pan-creas is identified while performing an extensive Kocher

maneuver The plane anterior to the SMV is developed under

the neck of the pancreas The neck of the pancreas is then

divided The specimen now remains connected by the head

and uncinate process of the pancreas These structures are

separated from the portal vein, SMV, and SMA With these

areas dissected, the specimen is removed and the pancreatic

neck margin, uncinate margin, bile duct margin, and

duo-denal or gastric margin are analyzed by intraoperative frozen

section to confirm that they are free of tumor

There are multiple options for reconstruction after

pan-creaticoduodenectomy Most commonly the

reconstruc-tion first involves the pancreas, followed by the bile duct,

ant then the duodenum The issues and controversies

sur-rounding the pancreatic and biliary reconstruction are

out-lined by multiple papers specifically addressing these issues

In brief, the pancreatic anastomosis can be performed to

the jejunum or to the stomach If the jejunum is used for

reconstruction, some groups favor a separate Roux-en-Yreconstruction for pancreas or even a double Roux-en-Yreconstruction for the pancreas and bile duct Controversycontinues regarding the best type of pancreaticojejunos-tomy, the importance of duct-to-mucosa sutures, and theuse of pancreatic duct stents At the Johns HopkinsHospital, the pancreatic reconstruction is typically per-formed with an end-to-end or end-to side pancreaticoje-junostomy to the proximal jejunum brought through adefect in the mesocolon to the right of the middle colicartery The biliary anastomosis is typically performed withan-end-to-side hepaticojejunostomy approximately 10 to

15 cm distal from the pancreaticojejunostomy If thepatient has a percutaneous biliary stent, then this is left inplace, traversing the anastomosis The third anastomosisperformed is the duodenojejunostomy in cases of pyloruspreservation, or the gastrojejunostomy in patients whohave undergone classic pancreaticoduodenectomy Thisanastomosis is typically performed downstream from thehepaticojejunostomy, either proximal or distal to the seg-ment of jejunum traversing the defect in the mesocolon.Figure 141-1 depicts the resection specimen and recon-struction after a pylorus preserving and classic pancreati-coduodenectomy After reconstruction is completed, closedsuction drains are left in place to drain the biliary and pan-creatic anastomosis Some groups prefer not to place closedsuction drains, accepting that if a fluid collection becomesclinically evident postoperatively, percutaneous drainage

by interventional radiology may be required

Resected

Resected

Retained Retained

FIGURE 141-1 Classic pancreaticoduodenectomy (A) and pylorus preserving pancreaticoduodenectomy (B) From Yeo and Cameron, 1988.

816 / Advanced Therapy in Gastroenterology and Liver Disease

Distal Lesions

Staging with laparoscopy is often of benefit with patients

with distal pancreatic cancers If metastatic disease is found,

distal pancreatectomy and splenectomy are unlikely to help

in the palliation of the patient Exposure for a distal

pan-createctomy and splenectomy can be obtained through a

vertical midline incision, or alternatively, a bilateral

sub-costal incision The spleen can technically be preserved for

benign disease, however, for cancer most groups prefer to

remove the spleen en bloc to gain wider margins and

incor-porate the lymph nodes in the splenic hilum The spleen is

mobilized towards the midline by dividing the lienorenal

ligament The short gastric vessels in the lienogastric

liga-ment are also divided A plane is then developed behind

the pancreatic tail and body, also mobilizing the splenic

artery and vein This dissection is continued several

cen-timeters beyond the tumor The splenic artery and vein are

isolated and suture ligated The body or tail of the pancreas

is then divided after placing a row of overlapping “U”

stitches in the remnant A frozen section is performed on

the pancreatic margin to confirm clearance of the lesion

and a closed suction drain is generally left in place in case

of leak from the pancreatic duct Several centers are now

beginning to use laparoscopic techniques for distal lesions,

especially if they have benign characteristics

Palliative

Unfortunately, only a minority of patients with AC of the

pancreas are suitable for resection and potential cure at

presentation Optimal palliation of symptoms to maximize

remaining quality of life is of primary importance to most

patients Nonoperative palliation is generally readily

avail-able in most centers The three main components of

pal-liation of unresectable periampullary ACs include (1)

drainage of the biliary tree, (2) relief of gastric outlet

obstruction, and (3) pain control The biliary tree may be

drained internally or externally with endoprostheses, as well

as metallic wall stents that can be inserted endoscopically

or percutaneously Some patients require percutaneous

access to the biliary tree and may be best palliated via a

per-cutaneous transhepatic drain that can be externalized when

obstructed There is a chapter on biliary tract endoscopy

(see Chapter 134, “Endoscopic Management of Bile Duct

Obstruction and Sphincter of Oddi Dysfunction”) New

approaches for palliative endostenting of gastric outlet

obstruction are now being tried with variable results

Chemical celiac splanchnicectomy may be performed via a

transcutaneous approach and is sometimes quite effective

at relieving the pain of locally invasive pancreatic cancer

Some centers also rely on surgical palliation Surgical

palliation offers the only chance for long term palliation of

the three major symptoms of periampullary AC Biliary

bypass can be performed with hepaticojejunostomy Unlike

endoscopic and percutaneous approaches, biliary bypassmay be more durable In a series reported from JohnsHopkins, recurrent jaundice developed in only 2% ofpatients receiving a palliative biliary bypass prior to death.(Lillemore et al, 1993) A second reason that some favor sur-gical palliation is that nonoperative palliation is frequentlyassociated with late complications of gastric outlet obstruc-tion In a prospective randomized trial of performing aprophylactic gastrojejunostomy in patients with unre-sectable periampullary AC, 19% of patients not undergo-ing gastrojejunostomy developed late duodenal obstructionrequiring intervention (Lillemore et al, 1999) The finalmajor advantage of operative palliation is the management

of pain A prospective randomized study has demonstratedthat intraoperative celiac axis injection with 50% alcoholcan both successfully relieve pain in patients with pain andprevent the development of pain in patients without pain

at the time of exploration (Lillemore et al, 1993)

Postoperative Results

During the 1960s and 1970s, many centers reported ative mortality rates following pancreaticoduodenectomy

oper-in the 20 to 40% range, with postoperative morbidity rates

as high as 40 to 60% During the last two decades, a matic decline in operative morbidity and mortality follow-ing this operation has been reported at a number of centers,with operative mortality rates in the range of 2 to 3% Somecenters have reported large series in excess of 100 patientswithout 1 perioperative death (Gilsdorf and Spanos, 1973).These dramatic improvements might be attributed to con-centration of these patients in high volume centers withfewer but more experienced surgeons performing the oper-ation and improved perioperative care (Sosa et al, 1997).Unfortunately, complications rates following pancreatico-duodenectomy remain high, usually in excess of 25 to 35%.Pancreatic fistula remains the most common serious com-plication following pancreaticoduodenectomy, with inci-dence raging from 5 to 15% (Lillemore et al, 1993) Theoverall mortality associated with pancreatic fistula hasgreatly diminished over the last several decades, thanks toimproved management The most frequent complicationfollowing pylorus-preserving pancreatic resection is delayedgastric emptying, with incidence ranging from 5 to 20% In

dra-most cases, delayed gastric emptying is temporary and

resolves spontaneously after a variable period of time,resulting in a delay in hospital discharge

Long-Term Survival

Survival following pancreaticoduodenectomy for ampullary carcinoma is highly dependent on the tumor’ssite of origin For instance, survival after resection of dis-tal bile duct, ampullary, and duodenal AC has always beensignificantly greater that that for pancreatic AC, with 5-year

peri-survival rates ranging from 30 to 50% (Yeo et al,1997) In

contrast, 5-year survival rates for patients with AC in the

head of the pancreas managed by

pancreaticoduodenec-tomy have been reported to reach 21% A number of

vari-ables have been evaluated by univariate and multivariate

analyses in an attempt to identify other factors predictive

of long-term survival Tumor characteristics found by

most investigators to be important predictors of survival

include tumor size, lymph node status, and resection

mar-gin status

Survival following pancreaticoduodenectomy for

peri-ampullary lesions that are not AC are strongly dependent on

the histology and biologic behavior of the primary lesion

resected Benign lesions that are completely resected

gener-ally result in very high long-term survival Neuroendocrine

tumors of the pancreas that are resected may have a variable

and somewhat indolent course

Supplemental Reading

Ahrendt GM, Demeure MJ Pancreatic islet cell tumors

exclud-ing gastrinoma In: Cameron JL, editor Current surgical

ther-apy, 7th ed St Louis: Mosby; 2001.

Gilsdorf RB, Spanos P Factors influencing morbidity and tality in pancreatico-duodenectomy Ann Surg 1973;177:332–7 Lillemoe KD, Cameron JL, Hardacre JM, et al Is prophylactic gas- trojejunostomy indicated for unresectable perimapullary cancer?

mor-A prospective randozimed trial mor-Ann Surg 1999;230:322–30 Lillemoe KD, Cameron JL, Kaufman HS, et al Chemical splanch- nicectomy in patients with unresectable pancreatic cancer A prospective randomized trial Ann Surg 1993;217:447–57 Lillemoe KD, Sauter PK, Pitt HA, et al Current status of surgi- cal palliation of periampullary carcinoma Surg Gynecol Obstet 1993;176:1–10.

Lillemoe KD, Yeo CJ, Cameron JL Pancreatic cancer: art care CA Cancer J Clin 2000;50:241–68.

state-of-the-Sosa JA, Bowman HM, Bass EB, et al Importance of hospital volume in the surgical management of pancreatic cancer Surg Forum 1997;48:584–6.

Yeo DJ, Cameron JL The pancreas In: Jardy JD, editor Hardy’s textbook of surgery, 2nd ed Philadelphia: JB Lippincott; 1988.

p 717–8.

Yeo CJ, Cameron JL, Lillemoe KD, et al Pancreaticoduodenectomy for cancer of the head of the pancreas 201 patients Ann Surg 1995;221:721–33.

Yeo DJ, Cameron JL, Sohn TA, et al Six hundred fifty tive pancreaticoduodenectomies in the 1990’s: Pathology, complications, outcomes Ann Surg 1997;226:248–60.

CHAPTER 142

DANLAHERU, MD

fractions d1-35) and chemotherapy (5-FU 200 mg/m2/d

as continuous infusion, weekly cisplatin 30 mg/m2venous [IV] bolus, and interferon [IFN]-α 3 million unitssubcutaneously every other day) during radiation orGITSG type chemotherapy with radiation therapy.Following combined modality chemoradiotherapy,chemotherapy alone was administered (5-FU 200 mg/m2/d

intra-as a continuous infusion) in two 6-week courses during

weeks 9 to 14 and 17 to 22 There were significant grade 3 and 4 gastrointestinal (GI) toxicities, including vomiting,

mucositis, diarrhea, and GI bleeding in the IFN-based

chemotherapy, requiring hospitalization in 43% of patients.

However, the majority of patients were still able to receive

>80% of planned therapy The median survival and 2-year survival were 46 months and 53% respectively for the IFN-

based chemoradiotherapy (Picozzi et al, 2003) TheAmerican College of Surgery Oncology Group (ACOSOG)will coordinate a similar multi-institutional phase II trial

in patients with pancreatic AC who are candidates forresection that has begun accrual in December 2003

In July 2002, the Radiation Therapy Oncology Group(RTOG) closed R97-04 This phase III trial randomized

518 resected PC patients to 5-FU continuous infusion(250 mg/m2/d for 3 weeks), followed by 5-FU continuousinfusion (250 mg/m2/d) during radiation therapy (50.4 Gy

in 1.8 Gy/fractions), followed by 2 cycles 5-FU continuous

The challenges in managing patients with pancreatic

can-cer (PC) are underscored by the seemingly immutable

sur-vival data, including a 5-year sursur-vival of 15 to 20% with a

median survival of 15 to 19 months for resectable disease

and 3% survival for all stages combined For patients with

locally advanced unresectable disease, median survival is 6

to 10 months, and for patients with metastatic disease it

is 3 to 6 months This chapter will describe the current

treatment recommendations as well as highlight the most

recent therapy advances for resected and advanced disease

Therapy for Adjuvant Disease

The current standard of 5-fluorouracil (5-FU) based

com-bined modality chemoradiotherapy is originally based on

data from the Gastrointestinal Tumor Study Group

(GITSG) This study was the first to document that

adju-vant therapy following surgical resection for pancreatic

surgery prolonged survival (Kalser and Ellenberg, 1985)

A number of groups have further developed this approach

and have, in general, also used 5-FU based chemotherapy

(Table 142-1)

Recently the Virginia Mason Medical Center published

their experience of 53 patients with resected pancreatic

adenocarcinoma (AC) who received combined

radiother-apy (external beam at a dose of 45 to 54 Gy in standard

TABLE 142-1 Selected Adjuvant Studies in Pancreatic Cancer

additional chemo

CI = continuous infusion; DPM = dipyridamole; 5-FU = 5-fluorouracil; IFN = interferon- α ; N/A = not applicable; NR = not reported; MMC = mitocycin; tx = treatment.

infusion, versus Gemcitabine 1000 mg/m2weekly for 3

weeks, followed by 5-FU continuous infusion during

radi-ation therapy, followed by 3 cycles Gemcitabine alone The

experimental question being asked was whether

Gemcitabine before and after 5-FU based

chemoradio-therapy would be more efficacious than continuous

infu-sion 5-FU before and after the same 5-FU based

chemoradiotherapy In 1997, when this study was designed,

there was inadequate knowledge regarding how to safely

administer Gemcitabine concurrently with irradiation to

allow for concurrent Gemcitabine and radiotherapy This

study was the first North American Co-operative group

trial since the GITSG trial Although the survival results

for this trial will not be known until 2004, a number of

important observations have already been made These

include that neither arm was associated with unacceptable

acute toxicity during the trial, that accrual was quite rapid

(12 to 14 patients per month) reflecting both the support

of the Eastern Co-operative Oncology Group (ECOG) and

the Southwest Oncology Group, and the willingness of

patients and their physicians to participate in adjuvant

tri-als for PC

Inconclusive Study Results

Despite a growing body of literature seemingly supporting

the benefit of adjuvant combined modality therapy

fol-lowing potentially definitive resection in patients with high

risk for recurrence, adjuvant chemoradiation has not been

universally accepted as standard of care A major criticism

has been that none of these studies included an

observa-tion only arm

There have now been two additional major studies that

have demonstrated inconsistent or negative conclusions

A European Organization for Research and Treatment of

Cancer (EORTC) trial randomized 218 patients with

pan-creatic and nonpanpan-creatic periampullary AC 2 to 8 weeks

following potentially curative resection to either

observa-tion or to combined radiotherapy (40 Gy using a 3 or 4

field technique in 2 Gy fractions with a 2 week break at

mid-treatment) and chemotherapy (5-FU administered as

a continuous infusion 25 mg/kg/d during the first week

of each 2 week radiation therapy module only) No

post-radiation chemotherapy was administered Median

pro-gression-free survival was 16 months in the observation

arm versus 17.4 months in the treatment arm (p = 643).

Median survival was 19 months in the observation group

versus 24.5 months in the treatment group, but was not

statistically significant (p = 737) For the subgroup of

patients with pancreatic AC (n = 114), the median survival

was 12.6 months in the observation group versus 17.1

months in the treatment arm but was not statistically

sig-nificant (p = 099) Of note, 21 of 104 patients randomized

to the treatment arm were not treated In addition,

although the original dose of 5-FU was already modest, 35

patients in the treatment arm received only 3 days of 5-FUduring the second module of radiotherapy secondary tograde 1 and 2 toxicities (Klinkenbijl et al, 1999) Therefore,this study could be better described as an underpoweredpositive study (see Table 142-1)

Recently, the European Study Group for PancreaticCancer (ESPAC) randomized 541 patients with pancreatic

AC in a 4 arm design based on a 2 ×2 factorial design,which included the following:

1 Observation

2 Concomitant chemoradiotherapy alone (20 Gy in 10fractions over 2 weeks with 500 mg/m25-FU IV bolusduring the first 3 days of radiation therapy); the mod-ule is repeated after a planned 2-week break, followed

by no additional chemotherapy

3 Chemotherapy alone (leukovorin 20 mg/m2bolus lowed by 5-FU 425 mg/m2administered for 5 consec-utive days repeated every 28 days for 6 cycles)

fol-4 Chemoradiotherapy followed by chemotherapy

There was no significant difference in survival between

patients assigned to chemoradiotherapy (median survival15.5 months) versus observation (median survival 16.1

months, p = 24) The survival data was similar in the set (n = 285 patients) randomized through the 2 ×2 design

sub-In contrast, there was a survival advantage for those patients treated with chemotherapy alone (median survival 19.7

months) versus the observation arm (median survival 14

months, p = 0005) For the same subset randomized

through the original 22 design, survival demonstrated atrend towards survival for chemotherapy alone (mediansurvival 17.4 months) versus observation alone (15.9

months) but was not statistically significant (p = 19).

Multivariate analysis for known prognostic factors ing margin status, lymph node involvement, and tumorgrade and size did not alter the effect for chemoradiother-apy treatment The study authors concluded that there was

includ-no survival benefit for adjuvant chemoradiotherapy In

addition, the authors concluded that a potential benefit

existed for adjuvant chemotherapy alone following cal resection (Neoptolemos et al, 2001) Although this was

surgi-a rsurgi-andomized study consisting of over 500 psurgi-atients, the

conclusions of the study should be carefully measured To

encourage maximal patient recruitment, the study wasmodified in that 68 patients were assigned separately andrandomized to either chemoradiotherapy or observation

In addition, 188 patients were subsequently assigned arately and randomized to either chemotherapy alone orobservation In a sense, three randomizations were possi-ble for inclusion into the same study Also, patients in theadditional two randomizations could have potentiallyreceived “background chemotherapy or chemotherapy”which was not specifically defined The background treat-ment was not known in 82 eligible patients Of note, thesepatients were still assigned into an arm of the study despite

sep-820 / Advanced Therapy in Gastroenterology and Liver Disease

lack of definitive knowledge of prior therapy Finally, 25 of

the eligible 541 patients refused to accept their

random-ization and an additional 25 patients withdrew secondary

to treatment toxicities

As the debate continues, there are several studies that

have recently opened or have been proposed by either the

cooperative groups or through single institutions Table

142-2 summarizes open or planned studies in the adjuvant

setting These future studies will be characterized by the

addition of multi-agent chemotherapy to irradiation at the

cooperative group level, by the addition of Gemcitabine to

the period of chemoradiation and by the use of conformal,

three-dimensional irradiation planned to patient specific

anatomic and surgical pathologic data

CURRENTPRACTICE

Nonetheless, most practitioners in the United States

employ radiation therapy (typically 54 Gy in 1.8 Gy

frac-tions) with simultaneous chemotherapy, the standard being

5-FU Although 5-FU can be administered in a number

of different schedules, most practitioners choose either

continuous infusion 200 to 250 mg/m2/d during radiation

therapy, or 500 mg/m2bolus given on the first 3 days and

last 3 days of radiation There is some interest in

substi-tuting an oral formulation of 5-FU known as Capecitabine

for continuous infusion 5-FU Although there is

prelimi-nary data primarily from the rectal cancer literature

demonstrating that Capecitabine can be safely combined

with radiation therapy, the comparison studies in PC have

not been completed

Role of Neoadjuvant Therapy

POTENTIALADVANTAGES

Neoadjuvant therapy is a potentially attractive alternative

to current standard adjuvant chemoradiation for severalreasons, including the following:

1 Radiation is more effective on well-oxygenated cellsthat have not been devascularized by surgery

2 Contamination and subsequent seeding of the toneum with tumor cells secondary to surgery couldtheoretically be reduced

peri-3 Patients with metastatic disease on restaging ing adjuvant therapy would not need to undergodefinitive resection and might benefit from palliativeintervention

follow-4 The risk of delaying adjuvant therapy would be inated because it would be delivered in the neoadju-vant setting

elim-There is significant published data primarily from MDAnderson Cancer Center and Fox Chase Cancer Centerusing chemoradiotherapy in a neoadjuvant approach forresectable PC To date, the current data demonstrate thatalthough neoadjuvant chemoradiotherapy can be admin-istered safely, there is no clear advantage to this strategywhen compared with postoperative therapy In patientswith marginally resectable disease, it remains to be seenwhether there is a meaningful cohort of patients for whomthis approach may represent an important therapeuticadvantage based on “downstaging” and improved surgi-cal outcomes

Currently, the ECOG is planning to open a prospectiverandomized trial randomizing patients to intensifiedGemcitabine based or Gemcitabine/5-FU/platinum basedchemoradiotherapy This trial makes an important dis-tinction between clearly unresectable disease and poten-tially resectable disease, especially around the issues ofpartial versus complete encasement of the superior mesen-teric artery and the length of superior mesenteric veininvolved by tumor at initial presentation

Treatment of Locally Advanced DiseasePancreatic tumors frequently invade adjacent structures,such as superior mesenteric and celiac vascular structures,making curative resection difficult if not impossible TheMemorial Sloan Kettering group recently reviewed theirexperience of 163 patients with locally advanced PC Anumber of chemotherapy regimens were integrated withradiation therapy and administered to 87 patients Onlythree patients had sufficient radiographic response to jus-tify surgical exploration Of these selected patients, one

of these underwent resection for curative intent (Kim et al,2002) For the approximately 30 to 40% of PC patients whopresent with such locally advanced, nonmetastatic disease,

TABLE 142-2 Active or Planned Adjuvant or

Neoadjuvant Studies

fx/Gem 600 mg/m 2 weekly followed by

gem for 3 cycles

ACOSOG Z05031 EBRT (50 Gy/5-FU CI/cisplatin/IFN, 5-FU CI

for 2 cycle)

Johns Hopkins GM-CSF allo vaccine, 5-FU CI, 5-FU CI/XRT,

5-FU CI for 2 cycles followed by GM-CSF

weekly for 6 wks with EBRT 50.4 Gy

followed by surgery, gem 1000 mg/m 2

over 100 min for 5 cycles

Arm B: Gem 175 mg/m 2 over 30 min day

1, 5, 28, 33/ cisplatin 20 mg/m 2 days

1–4, 29–33, 5-FU 500 mg/m 2 over

21 hrs days 1–4, 29–32 followed by

EBRT 50.4, surgery, gemzar for 3 cycles

allo = allogeneic; CI = continuous infusion; 5-FU = 5-fluorouracil; Gem = gemcitabine; GM-CSF =

granulocyte macrophage colony stimulating factor; IFN = interferon- α ; LV = leukovorion; MMC =

optimal management is controversial Palliative surgery,

chemoradiation, chemotherapy alone, and locally directed

therapies have all been employed in this setting

Metastatic Disease

In patients with metastatic disease, the current standard of

care is single agent Gemcitabine Burris and colleagues

(1997) randomized 126 patients with unresectable PC to

either Gemcitabine (1000 mg/m2weekly over a 30 minute

infusion 7 times followed by 1 week rest then weekly 3 times

every 4 weeks) or 5-FU (600 mg/m2weekly) Although the

primary endpoints were issues related to quality of life

(per-formance status, weight gain, analgesic consumption, and

pain), median survival was 5.7 months in the Gemcitabine

arm compared with 4.4 months in the 5-FU arm In

addi-tion, 1-year survival was 18% in the Gemcitabine arm

com-pared with 2% in the 5-FU arm (p = 0025) with median

time to progression also favoring Gemcitabine (9 weeks

compared with 4 weeks in the 5-FU arm, p = 0002).

Gemcitabine was well tolerated with the majority of side

effects related to grade 3 or 4 neutropenia (26%) without

associated infections, low grade fevers (30%), and nausea

and vomiting (9.5% and 3.2%) (Burris et al, 1997) Based

on this study, Gemcitabine was approved for the treatment

of patients with advanced PC in the United States and many

other countries and is currently considered the standard

agent for the treatment of this disease as well as the accepted

control with which to compare new drugs and interventions

Recent efforts have focused on developing strategies

that would enhance the efficacy of Gemcitabine and

ulti-mately improve on median survival These strategies

include identifying alternative dosing schedules of

Gemcitabine that might enhance drug delivery to tumor

cells, as well as identifying synergistic combinations with

other chemotherapeutic agents Tempero and colleagues

(2003) randomized 92 patients to either Gemcitabine

(2200 mg/m2) over the standard 30-minute infusion or

Gemcitabine (1500 mg/m2) at a rate of 10 mg/m2/min Thedrug was given weekly for 3 consecutive weeks every 4 weeks

in both arms of the study Patients treated with the dose-rate regimen experienced more toxicity, with a 49 and37% occurrence of neutropenia and thrombocytopenia ver-sus a 28 and 10% occurrence, respectively, in patients treated

fixed-in the conventional schedule (Tempero et al, 2003) Patients

on the fixed-dose-rate had a higher response rate (11.6 vs

4.1 %), median survival (8 vs 5 months, p = 013) and 1-year

survival (23.8 vs 7.3 %) than patients treated on the ventional schedule This strategy is now being tested in ran-domized phase III studies

con-Other potentially synergistic agents that have been usedwith Gemcitabine include Cisplatin, Irinotecan, 5-FU,antifolates, such as raltitrexed and pemetrexed, andTaxotere and Oxaliplatin Some of these studies are high-lighted in Table 142-3 Although some of these studiesappear promising, the results are preliminary

New Drugs in PC

During the last few years, an increasing number of newdrugs, many of them targeted to specific alterations inmalignant cells, have been tested in PC The rationale todevelop these drugs in PC comes from the better under-standing of the biological basis of the disease that has madepossible the identification and validation of some of thesetargets in PC In addition, the poor prognosis of patientswith this disease and the evidence from clinical trials dis-cussed above that conventional chemotherapy may havereached a plateau with regards to improving outcome hasalso motivated an aggressive evaluation of new drugs in

PC To date, targeted drugs such as the matrix proteinase inhibitors (marimastat and Bay12-9566),inhibitors of angiogenesis (bevacizumab), agents targeted

metallo-to the ras oncogene (R115777 and Lonafarnib), and

inhibitors of the epidermal growth factor receptor (EGFR)family of membrane receptors (trastuzimab, ceuximab),

TABLE 142-3 Selected Studies in Advanced Pancreatic Cancer

5-FU = 5-fluorouracil; Gem = gemcitabine; NR = no response; PR = partial response

822 / Advanced Therapy in Gastroenterology and Liver Disease

immunotherapy and gene therapy have been evaluated in

this patient population with mixed results

SummaryGemcitabine is currently the only approved chemothera-

peutic agent that has demonstrated significant antitumor

effects in advanced PC Although the efficacy of

Gemcitabine may be augmented by innovative dosing

schedules or by the use of synergistic drug combinations,

the standard regimen to date remains single agent

Gemcitabine This strategy is also appropriate for patients

with locally advanced disease, though these patients are

commonly managed with combined modality approaches

Either a conventional 30-minute or fixed-dose-rate

infu-sion is appropriate based on existing data Combinations

of Gemcitabine with other agents such as cisplatin,

irinote-can, oxaliplatin, and fluoropyrimidines have not

consis-tently resulted in significant improvement in survival or

quality of life in studies available thus far and should not

be considered the standard of care at the present time;

how-ever, this could change as the results of randomized

stud-ies are available Given the data with conventional

treatments, enrollment in a clinical trial should still be the

preferred approach to these patients

Supplemental Reading

Burris HA, Moore MJ, Anderson J, et al Improvements in survival

and clinical benefit with gemcitabine as first-line therapy for

patients with advanced pancreas cancer: a randomized trial J

Clin Oncol 1997;15:2403–13.

Fine RL, Sherman W, Chabot J, et al Biochemically synergistic

chemotherapy for advanced pancreatic cancer [abstract 575].

Proc ASCO 2002;21:145a.

Heinemann V, Quietzsch D, Gieseler F, et al A phase III trial comparing gemcitabine plus cisplatin vs gemcitabine alone in advanced pancreatic carcinoma [abstract 1003] Proc ASCO 2003;22:250.

Kalser MH, Ellenberg SS Pancreatic cancer: adjuvant combined radiation and chemotherapy following curative resection Arch Surg 1985;120:899–903.

Kim HJ, Kzischke K, Brennan MF, et al Does neoadjuvant chemoradiation downstage locally advanced pancreatic cancer? J Gastrointest Surg 2002;6:763–9.

Klinkenbijl JH, Jeekel J, Sahmoud T, et al Adjuvant radiotherapy and 5-Fluorouracil after curative resection of cancer of the pancreas and periampullary region Ann Surg 1999;230:776–84 Louvet C, Labianca R, Hammel P, et al Gemcitabine versus GEMOX (Gemcitabine + oxaliplatin) in nonresectable pancreatic adenocarcinoma: interim results of the GERCOR/GISCAD Intergroup Phase III [abstract 1004] Proc ASCO 2003;22:250 Neoptolemos JP, Dunn JA, Stocken DD, et al Adjuvant chemoradiotherapy and chemotherapy in resectable pancreatic cancer: a randomized controlled trial Lancet 2001;358:1576–85 Picozzi VJ, Kozarek RE, Jacobs AD, et al Adjuvant therapy for resected pancreas cancer (PC) using alpha-interferon (IFN)- based chemoradiation: completion of a phase II trial [abstract 1061] Proc ASCO 2003;22:265.

Rocha Lima CMS, Rotche R, Jeffery M, et al A randomized phase III study comparing efficacy and safety of gemcitabine (GEM) and Irinotecan (I), to gemcitabine (GEM) alone in patients with locally advanced or metastatic pancreatic cancer who have not received prior systemic therapy [abstract 1005] Proc ASCO 2003;22:251.

Sohn TA, Yeo CJ, Cameron JL, et al Resected adenocrcinoma of the pancreas-616 patients: results, outcomes, and prognostic indicators J Gastrointest Surg 2000;4:567–79.

Tempero M, Plunkett W, van Haperan VR, et al Randomized phase II comparison of dose-intense Gemcitabine: thirty- minute infusion and fixed dose rate infusion in patients with pancreatic adenocarcinoma J Clin Oncol 2003:21:1–7.

lives could be saved The large size of most noninvasive

IPMNs (3 cm in some series), suggests that it should be

possible to detect these neoplasms using conventional

imaging techniques such as endoscopic ultrasound (EUS)

or computerized tomography Indeed, we have recently

reported an asymptomatic patient who was found to have

a noninvasive IPMN by EUS He underwent EUS because

of a history that suggested he was at increased risk for

developing a pancreatic neoplasm (see later section on

familial PC) The IPMN was resected and this patient is

presumably cured before a life threatening invasive

can-cer developed

MCNs

MCNs of the pancreas are distinctive cystic neoplasms that

arise primarily (90%) in women In contrast to IPMNs,

which usually arise in the head of the gland and in the larger

pancreatic ducts, MCNs usually arise in the tail of the gland.

MCNs almost never connect to the larger pancreatic ducts.

The mean age at diagnosis is between 40 and 50 years (range

14 to 95 years) and most patients present with nonspecific

symptoms including epigastric pain or a sense of

abdomi-nal fullness These patients can also develop

gastrointesti-nal (GI) symptoms, including nausea and vomiting,

diarrhea, anorexia, and weight loss Remarkably, as many

as 20% of MCNs are discovered incidentally during

abdom-inal imaging for unrelated indications This percentage can

be expected to increase with the growing use of screening

imaging studies in asymptomatic individuals

Pathologically, MCNs are usually large (mean 7 to

10 cm) cystic masses filled with thick tenacious mucin In

contrast to IPMNs, these neoplasms typically do not

con-nect with the larger pancreatic ducts, and therefore these

patients do not have mucin oozing from the ampulla of

Vater MCNs are lined by a columnar mucin-producing

epithelium, and, of note, they have a dense underlying

stroma that resembles the stroma seen in the ovary

(“ovar-ian stroma”) Just as was true for IPMNs, MCNs can show

varying degrees of atypia Noninvasive MCNs without

sig-nificant cytologic or architectural atypia are designated

mucinous cystadenoma, those with moderate architectural

and/or cytologic atypia are designated MCNs with

mod-erate dysplasia, and those with significant architectural and

cytologic atypia are designated MCN-in situ carcinoma.

One-third of all MCNs are associated with an invasive AC.

These neoplasms are designated as mucinous

cystadeno-carcinomas with an associated invasive carcinoma The

inva-sive carcinomas are usually a tubular/ductal type of invainva-sive

AC The diagnosis of an invasive carcinoma is based on the

presence of tissue invasion by the neoplastic cells Lymph

node metastases are identified in about one-fourth of all

surgically resected mucinous cystadenocarcinomas with

an associated invasive carcinoma

Surgery and Survival

Surgical resection is the treatment of choice for all MCNs.The 5-year disease-specific survival rate for patients withsurgically resected noninvasive MCNs is 100% If an inva-sive cancer is present, the 5-year disease specific survivalrate drops to 50% The survival rate for patients with unre-sectable mucinous cystadenocarcinomas with an associ-

ated invasive carcinoma is even worse (20% at 2 years) The goal of surgical resection of MCNs is complete resection of the neoplasm with negative margins It may be possible to pre-

serve the spleen in patients with left-sided tumors without

an invasive component

Just as is true for patients with IPMNs, the long tory of symptoms experienced by most patients withMCNs, the association of noninvasive MCNs with an inva-sive cancer, and molecular genetic analyses of MCNs, all

his-suggest that noninvasive MCNs can and do progress to sive cancer over time This has enormous clinical implica-

inva-tions It suggests that the lives of patients with MCNs can

be saved if their neoplasms can be detected and surgicallyresected before an invasive cancer develops

MCNs should be distinguished from serous cystadenomas

of the pancreas because the latter are almost always benign.

The presence of a central stellate scar and innumerable smallcysts should suggest the diagnosis of a serous cystadenoma.Pancreatic Intraepithelial Neoplasia

Pancreatic intraepithelial neoplasms (PanINs) are microscopic

lesions in the small pancreatic ducts and ductules Althoughthese lesions have been recognized for decades, their clini-cal importance as the likely noninvasive precursors to inva-sive AC of the pancreas has only recently been established.PanINs are not recognizable grossly Instead, these are

small microscopic lesions PanINs can show varying degrees

of atypia, and a system to classify these lesions has recently

been adopted PanIN-1A is the designation given to flat lesions without significant atypia and PanIN-1B to papil-

lary lesions without atypia Lesions with moderate

dys-plasia are designated PanIN-2, and those with significant architectural and cytologic atypia are designated PanIN-3

(carcinoma in situ)

Three lines of evidence strongly suggest that PanINs arethe precursors to invasive AC of the pancreas First, PanINsare often found in the pancreatic parenchyma adjacent to aninvasive AC Second, there have been several anecdotal casereports of patients with histologically proven PanINs wholater develop an invasive AC Third, molecular genetic analy-ses of PanINs have shown that they harbor many of the samegenetic alterations found in invasive pancreatic ACs.Therefore, just as there is a progression in the colorectum

from adenoma to invasive carcinoma, so too do we believe that there is a progression in the pancreas from PanIN-1 to PanIN-2 to PanIN-3 and eventually to invasive AC This has

Neoplastic Cysts and Other Precancerous Lesions of the Pancreas / 825

enormous clinical potential, considering the almost

uni-formly fatal outcome of invasive AC of the pancreas We also

know from autopsy studies of patients without known

pan-creatic disease that PanINs occur with increasing frequency

with advancing age, but in the absence of an invasive cancer

these are almost always low grade PanINs (PanIN-1)

Clearly, if techniques can be developed to prevent,

detect, and treat PanINs, particularly high grade PanINs,

before they develop into an invasive AC, then lives will be

saved Unfortunately, PanINs are extremely small lesions;

they are significantly smaller than the resolution of most

currently available imaging modalities A great deal of

effort is therefore going into the development of

molecu-lar based screening/diagnostic tests that have the potential

to detect neoplasms as small as PanINs

At-Risk Publications

Clearly, as exciting as the characterization of each of these

noninvasive precursor lesions in the pancreas is, the

chal-lenge is to define subgroups in the population who might

best benefit from screening for early pancreatic neoplasia

Individuals with a strong family history of PC and

indi-viduals with certain genetic syndromes may represent a

subgroup at great enough risk to warrant screening

Familial PC

Although the familial aggregation of other forms of cancer

has been recognized for decades, the importance of family

history as a risk factor for PC has become clearer in the past

several years Individuals with a strong family history of PC

have an increased risk of developing PC themselves

Tersmette and colleagues (2001) studied kindreds enrolled

in the National Familial Pancreas Tumor Registry at Johns

Hopkins <http://pathology.jhu.edu/PANCREAS_NFPTR/>

and found that healthy individuals in kindreds in which 3

or more family members had previously been diagnosed

with PC had a 56-fold increased risk of developing PC

themselves The risk of developing PC in these kindreds may

be particularly high among cigarette smokers Clearly,

because of their increased risk, individuals with a strong

family history of PC would be one of the first groups to

ben-efit from screening for early pancreatic neoplasia

Although the genetic basis for the aggregation of PC

in most kindreds has not yet been identified, five genetic

syndromes have been identified that increase the risk of PC

(Table 143-1) These include the following:

1 Second breast cancer gene (BRCA2) syndrome

In addition, recent studies by van der Heijden and

col-leagues (2003) have identified inherited mutations in the Fanconi anemia genes, FANCC and FANCG, in a small per- centage of patients with young age of onset PC.

The BRCA2 syndrome is caused by germline mutations in the BRCA2 gene and these patients not only have an increased

risk of developing breast, ovarian, and prostate cancer, butthey also have a 3.5 to 10-fold increased risk of developing

PC To date, germline mutations in the BRCA2 gene account

for 16% of the kindreds with an aggregation of PC

The Peutz-Jeghers syndrome is characterized by

pig-mented macules on the lips and buccal mucosa and tomatous polyps of the GI system Peutz-Jeghers is caused

hamar-by germline (inherited) mutations in the STK11/LKB1 gene

on chromosome 19 Giardiello and colleagues (2000) haverecently shown that patients with the Peutz-Jeghers syn-drome have a 132-fold increased risk of developing PC

The FAMMM syndrome is characterized by nevi,

atyp-ical nevi, melanomas, and a 20 to 34-fold increased risk

of PC It is caused by germline mutations in the p16 gene

on chromosome 9p

HNPCC is associated with an increased risk of

colorec-tal, endometrial, gastric, ovarian, small intestinal, ureter,and PC HNPCC is caused by germline mutations in one ofthe deoxyribonucleic acid (DNA) mismatch repair genes

(particularly hMLH1 and hMSH2), and, as a result, cancers

in patients with HNPCC are characterized by lite instability, the hallmark of a mismatch repair defect

microsatel-Familial (hereditary) pancreatitis is characterized by the

autosomal dominant inheritance and young age of onset

of pancreatitis Familial pancreatitis has been shown to becaused by germline mutations in the cationic trypsinogen

gene (PRSS1) and patients with familial pancreatitis have

a 50 to 80-fold increased risk of developing PC

IMPLICATIONS

The identification and characterization of well-definedgenetic syndromes associated with an increased risk of PChas several important implications First, most of these syn-dromes are also associated with an increased risk of othercancer types For example, FAMMM is associated with an

TABLE 143-1 Risks of Pancreatic Cancer

increased risk of both melanoma and PC The recognition

of a patient with FAMMM should therefore lead to

increased screening for melanoma, and this screening may

save lives

Second, in selected instances patients with one of these

syndromes may elect to undergo pancreatectomy For

example, older patients with hereditary pancreatitis may

have evidence of both pancreatic endocrine and exocrine

insufficiency, and yet harbor a 25 to 40% lifetime risk of

developing PC In patients with chronic abdominal pain

(narcotic dependent) and hereditary pancreatitis, total

pan-createctomy can be offered, to alleviate pain and reduce

cancer risk Additionally, it is not unreasonable to discuss

“prophylactic” total pancreatectomy for patients with

markedly increased risk, noting that patient education

about the consequences of pancreatectomy (endocrine and

exocrine insufficiency) is mandatory

Prophylactic total pancreatectomy is, however, a high risk

procedure with a high rate of long term morbidity and

mortality A third option, effective screening for early

pan-creatic neoplasia, is urgently needed Brentnall and

col-leagues, (1999) have suggested screening for pancreatic

dysplasia with EUS Abnormal EUS findings can be

sam-pled using fine needle aspiration and further investigated

with other imaging techniques, such as ERCP Patients with

suspicious cytology and imaging findings can then be

con-sidered for surgical resection Such an approach has been

shown to detect early noninvasive pancreatic neoplasms,

including IPMNs, and it is predicted to increase patient life

expectancy, hopefully in a cost effective manner.*

Future Screening Modalities

Because of the low prevalence of PC in the population, any

useful screening modalities will only target individuals at

high risk of developing PC, such as those with a strong

family history of PC and those with germline mutations in

one of the PC causing genes The main goals of screening

are to detect pancreatic neoplasia before invasive PC

devel-ops and, failing that, to detect asymptomatic PCs while they

are still resectable In this regard, using current imaging

tests such as EUS, experienced investigators can detect

prevalent small (1 to 2 cm) pancreatic neoplasms More

challenging is detecting microscopic high grade PanIN

lesions The detection of these microscopic lesions will

probably require molecular assays

Molecular assays will need to be able to distinguish

pan-creatic neoplasia from chronic pancreatitis For a

molec-ular assay to succeed as a screening test, it will need to be

applied to the appropriate clinical sample Apoptotic

pan-creatic carcinoma cells including tumor DNA and proteins

are released through the pancreatic ducts into pancreatic juice, duodenal fluid, stool, and, to a lesser extent, into the

blood Although all of these secondary sources can be pled, an ideal diagnostic marker of pancreatic neoplasiawould be measurable in serum Unfortunately, in the set-ting of early invasive PC, existing serum markers are oftennormal This suggests that in order to detect pre-invasivelesions of the pancreas, a more direct sampling of the pan-creas will be required Pancreatic juice can be collected dur-ing routine upper GI endoscopy after secretin stimulation.Higher levels of cancer DNA and proteins make pancreaticjuice a potentially optimal specimen to use when screen-ing high risk patients for PC, analogous to sputum for lungcancer or nipple aspirates for breast cancer

sam-The ideal marker or marker panel of pancreatic plasia has not yet been identified Potential biomarkers of

neo-PC can be divided into three biochemical targets based techniques aim to detect cancer specific DNA alter-ations, such as mutations in genomic DNA, CpG islandmethylation, and mutations in mitochondrial DNA.Ribonucleic acid-based detection methods have been used

DNA-to identify overexpressed genes in secondary fluids, andglobal gene expression profiles using oligonucleotidemicroarrays could be obtained from clinical samples, such

as fine needle aspirates of suspicious lesions or from creatic juice Finally, protein-based detection methods

pan-remain the mainstay of cancer markers Tumor markerscan be identified as proteins upregulated in cancer cellsusing mass spectrometry-based approaches, or by usingantibody-based methods to the protein product of genesknown to be overexpressed in cancer

Among DNA markers, because of their prevalence andease of detection, DNA methylation abnormalities havepotential for use in early detection strategies One limita-tion to DNA methylation as a marker of cancer is that theabnormal methylation changes that arise during neoplas-tic development in one tissue may be present in histolog-ically normal cells of adjacent tissues

Given the current limitations of DNA markers, an ideal

PC marker may be protein based Large-scale analyses of theproteins in biological fluids (known as “proteomics”) areunderway Thus, perhaps the most promising molecularstrategies for early PC detection and for detecting advancedPanINs are through proteomics profiling of the serum andpancreatic juice of individuals at high risk of developing PC

Supplemental Reading

Adsay NV, Longnecker DS, Klimstra DS Pancreatic tumors with cystic dilatation of the ducts: intraductal papillary mucinous neoplasms and intraductal oncocytic papillary neoplasms Semin Diagn Pathol 2000;17:16–30.

Brentnall TA, Bronner MP, Byrd DR, et al Early diagnosis and treatment of pancreatic dysplasia in patients with a family history of pancreatic cancer Ann Intern Med 1999;131:247–55.

*Editor’s Note: Investigators working with families with hereditary

pancreatitis have been considering surveying protocols for that

population.